Intelligent Methods Systems And Applications In Computing Communications And Control Simona Dzitac

Intelligent Methods Systems And Applications In
Computing Communications And Control Simona
Dzitac download
https://ebookbell.com/product/intelligent-methods-systems-and-
applications-in-computing-communications-and-control-simona-
dzitac-46364628
Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.
Evolving Connectionist Systems Methods And Applications In
Bioinformatics Brain Study And Intelligent Machines 1st Edition Nikola
Kasabov Phd
https://ebookbell.com/product/evolving-connectionist-systems-methods-
and-applications-in-bioinformatics-brain-study-and-intelligent-
machines-1st-edition-nikola-kasabov-phd-4200540
Advances In Hybridization Of Intelligent Methods Models Systems And
Applications 1st Edition Ioannis Hatzilygeroudis
https://ebookbell.com/product/advances-in-hybridization-of-
intelligent-methods-models-systems-and-applications-1st-edition-
ioannis-hatzilygeroudis-6790406
Intelligent Systems Applications In Software Engineering Proceedings
Of 3rd Computational Methods In Systems And Software 2019 Vol 1 1st Ed
2019 Radek Silhavy
https://ebookbell.com/product/intelligent-systems-applications-in-
software-engineering-proceedings-of-3rd-computational-methods-in-
systems-and-software-2019-vol-1-1st-ed-2019-radek-silhavy-10800248
Advances In Manufacturing Processes Intelligent Methods And Systems In
Production Engineering Progress In Application Of Intelligent Methods
And Systems In Production Engineering Andre Batako
https://ebookbell.com/product/advances-in-manufacturing-processes-
intelligent-methods-and-systems-in-production-engineering-progress-in-
application-of-intelligent-methods-and-systems-in-production-
engineering-andre-batako-50750808

Condition Monitoring Using Computational Intelligence Methods
Applications In Mechanical And Electrical Systems 1st Edition
Tshilidzi Marwala
https://ebookbell.com/product/condition-monitoring-using-
computational-intelligence-methods-applications-in-mechanical-and-
electrical-systems-1st-edition-tshilidzi-marwala-2516166
Intelligent Systems And Technologies Methods And Applications 1st
Edition Lakhmi C Jain
https://ebookbell.com/product/intelligent-systems-and-technologies-
methods-and-applications-1st-edition-lakhmi-c-jain-4193244
Computational Statistics And Mathematical Modeling Methods In
Intelligent Systems Proceedings Of 3rd Computational Methods In
Systems And Software 2019 Vol 2 1st Ed 2019 Radek Silhavy
https://ebookbell.com/product/computational-statistics-and-
mathematical-modeling-methods-in-intelligent-systems-proceedings-
of-3rd-computational-methods-in-systems-and-software-2019-vol-2-1st-
ed-2019-radek-silhavy-10800604
Cybernetics Approaches In Intelligent Systems Computational Methods In
Systems And Software 2017 Vol 1 Prokopova
https://ebookbell.com/product/cybernetics-approaches-in-intelligent-
systems-computational-methods-in-systems-and-
software-2017-vol-1-prokopova-6840142
Artificial Intelligent Methods For Handling Spatial Data Fuzzy
Rulebase Systems And Gridded Data Problems Jrg Verstraete
https://ebookbell.com/product/artificial-intelligent-methods-for-
handling-spatial-data-fuzzy-rulebase-systems-and-gridded-data-
problems-jrg-verstraete-7174876

Advances in Intelligent Systems and Computing1435
Simona Dzitac · Domnica Dzitac ·
Florin Gheorghe Filip ·
Janusz Kacprzyk ·
Misu-Jan Manolescu ·
Horea Oros Editors
Intelligent
Methods Systems
and Applications
in Computing,
Communications and
Control
9th International Conference
on Computers Communications and
Control (ICCCC) 2022

Advances in Intelligent Systems and Computing
Volume 1435
Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
Advisory Editors
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing,
Universidad Central de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, School of Computer Science and Electronic Engineering,
University of Essex, Colchester, UK
LászlóT. Kóczy, Department of Automation, Széchenyi István University,
Gyor, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas
at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao
Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information Technology,
University of Technology Sydney, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute
of Technology, Tijuana, Mexico
Nadia Nedjah, Department of Electronics Engineering, University of Rio de Janeiro,
Rio de Janeiro, Brazil
Ngoc Thanh Nguyen
, Faculty of Computer Science and Management,
Wrocław University of Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering,
The Chinese University of Hong Kong, Shatin, Hong Kong

The series“Advances in Intelligent Systems and Computing”contains publications
on theory, applications, and design methods of Intelligent Systems and Intelligent
Computing. Virtually all disciplines such as engineering, natural sciences, computer
and information science, ICT, economics, business, e-commerce, environment,
healthcare, life science are covered. The list of topics spans all the areas of modern
intelligent systems and computing such as: computational intelligence, soft comput-
ing including neural networks, fuzzy systems, evolutionary computing and the fusion
of these paradigms, social intelligence, ambient intelligence, computational neuro-
science, artiﬁcial life, virtual worlds and society, cognitive science and systems,
Perception and Vision, DNA and immune based systems, self-organizing and
adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics
including human-machine teaming, knowledge-based paradigms, learning para-
digms, machine ethics, intelligent data analysis, knowledge management, intelligent
agents, intelligent decision making and support, intelligent network security, trust
management, interactive entertainment, Web intelligence and multimedia.
The publications within“Advances in Intelligent Systems and Computing”are
primarily proceedings of important conferences, symposia and congresses. They
cover signiﬁcant recent developments in theﬁeld, both of a foundational and
applicable character. An important characteristic feature of the series is the short
publication time and world-wide distribution. This permits a rapid and broad
dissemination of research results.
Indexed by DBLP, INSPEC, WTI Frankfurt eG, zbMATH, Japanese Science and
Technology Agency (JST).
All books published in the series are submitted for consideration in Web of Science.
For proposals from Asia please contact Aninda Bose ([email protected]).
More information about this series athttps://link.springer.com/bookseries/11156

Simona DzitacDomnica Dzitac
Florin Gheorghe FilipJanusz Kacprzyk
Misu-Jan ManolescuHorea Oros
Editors
IntelligentMethodsSystems
andApplications
inComputing,
CommunicationsandControl
9th International Conference on Computers
Communications and Control (ICCCC) 2022
123

Editors
Simona Dzitac
University of Oradea
Oradea, Romania
Florin Gheorghe Filip
Romanian Academy
Member of the Romanian Academy
Bucharest, Romania
Misu-Jan Manolescu
Agora University of Oradea
Oradea, Romania
Domnica Dzitac
New York University Abu Dhabi
Abu Dhabi, United Arab Emirates
Janusz Kacprzyk
Polish Academy of Sciences
Systems Research Institute
Warszawa, Poland
Horea Oros
University of Oradea
Oradea, Romania
ISSN 2194-5357 ISSN 2194-5365 (electronic)
Advances in Intelligent Systems and Computing
ISBN 978-3-031-16683-9 ISBN 978-3-031-16684-6 (eBook)
https://doi.org/10.1007/978-3-031-16684-6
©The Editor(s) (if applicable) and The Author(s), under exclusive license
to Springer Nature Switzerland AG 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, speciﬁcally the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microﬁlms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional afﬁliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface
The volume contains a selection of 37 papers presented at the ninth edition of the
International Conference of Computers, Communications and Control-ICCCC. The
initial idea of the conference which was founded in 2006 by Ioan Dziţac, Florin
Gheorghe Filip and Mişu Jan Manolescu was to set the stage for presenting papers
containing scientiﬁc and practical results concerning the design and deployment of
systems that integrate the three technologies:computers,communicationsand
controlat large. It is worth noticing that the conference was intended by its
founders to be an opportunity offered to academia people and industry practitioners
to exchange ideas and discuss possible future common projects in a friendly
environment.
Since 2006, the conference has been organized every even year byAgora
University of Oradea, under the aegis of theInformation Science and Technology
Section of Romanian Academy. The published conference proceedings and website
witness a successful series of events.
From the very beginning, the conference was envisioned as a public and open
event associated with theInternational Journal of Computers, Communications and
Control(IJCCC), a publication which was meant to be a perennial means to archive
and broadly disseminate selected articles describing the results obtained in the
well-deﬁned domain of solutions integrating the three technologies.
Over the years, outstanding scholars and successful professionals from industry
presented high-quality papers, which generated the need of diversifying the set of
core topics to be addressed.Soft computing(with particular emphasis onfuzzy sets
and systemsand associated solvers of complex and complicated problems),
decision-making modelsandcomputer-based support systemshave found their
place in the sections of the conference. More recently,greenandtrustworthy
computinghas been accepted as a core topic of the conference in line with the
current developments in the international scientiﬁc and technological landscape.
Consequently, the ninth edition of the conference was organized intoﬁve sections
v

1. Integrated Solutions in Computer-Based Control;
2. Advanced Control Systems Integrating Computers and Communications;
3. Soft Computing Including Fuzzy System Approach;
4. Decision-Making and Support Systems;
5. Trustworthy and Green Design.
The ninth edition of ICCCC was dedicated in memory of Prof. Ioan Dziţac, the
conference founder and the general chair of the previous eight editions. On this
occasion, the Dziţac Family (wife Simona and daughter Domnica) awarded the
“Ioan Dziţac Prize”of EUR 500 for the most appreciated scientiﬁc work presented
at ICCCC 2022 by a promising young researcher, more precisely to the PhD student
of the University of Granada, Spain—JoséRamón Trillo. The conference general
chair was prof. Simona Dziţac.
The programme included four key note speeches,ﬁve invited talks and 41
regular papers grouped in the aboveﬁve sections. The papers were authored by
researchers from 18 countries (Canada, Chile, China, Colombia, Japan, Israel, Italy,
Lithuania, Mexico, Poland, Romania, Russia, Serbia, Spain, Switzerland, UAE, UK
and USA).
The keynote speeches were held by high reputation scholars such as
1. Acad. Gintautas Dzemyda, Vilnius University, Lithuania (Geometric
Multidimensional Scaling: A New Approach for Data Dimensionality
Reduction);
2. Prof. Enrique Hererra-Viedma, University of Granada, Spain (On Fuzzy and
Linguistic Decision-Making: Scenarios and Challenges);
3. Acad. Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland
(AI-enabled Decision Support Systems: An Effective And Efﬁcient Solution For
Complex Problems);
4. Prof. Gaston Lefranc H., Pontiﬁcia Universidad Católica de Valparaíso, Chile
(Review of Trends in Manufacturing Systems Based On Industry 4.0:
The Opportunities).
The invited speakers were
1. Prof. Razvan Andonie, Central Washington University, Ellensburg, USA
(Computational Semiotics in Deep Learning);
2. Dr. Valeriu Beiu,“Aurel Vlaicu”University of Arad, Romania (Reﬂective
Optics from Stars to Dust);
3. Prof. Yezid Donoso, Universidad de los Andes, Bogota, Colombia
(Methodologies for Solving Complex Multi-Objective Combinatorial Problems
in Engineering: An Evolutionary Approach Applied to Computer Networks);
4. Prof. Alfred Bruckstein, Technion, Haifa, Israel (Guiding Swarms with
Pheromone Gradients or Broadcast Signals);
5. Prof. Valentina E. Balas,“Aurel Vlaicu”University of Arad, Romania
(Advances in Soft Computing and Applications).
vi Preface

Fifty-three regular papers were submitted for evaluation. Forty-one papers were
accepted by theInternational Programme Committee. Five of the papers presented
at ICCCC 2022 have obtained the“Best paper award”. A number of 37ﬁnal
versions of the accepted papers are included in theﬁve sections of the volume as
follows:
Section 1.Integrated Solutions in Computer-Based Control: eight papers;
Section 2.Advanced Control Systems Integrating Computers and Communications:
seven papers;
Section 3.Soft Computing Including Fuzzy System Approach: eight papers;
Section 4.Decision-Making and Support Systems: nine papers;
Section 5.Trustworthy and Green Design:ﬁve papers
The study of the papers contained in the volume can be useful for researchers,
consultants and postgraduate students incomputer science and design,applied
informatics,automationandindustrial engineering.
The editors are grateful to Springer for accepting to publish the proceedings of
ICCCC in theAdvances in Intelligent Systems and ComputingSeries and ensuring a
high-quality technical version of the volume.
June 2022 The editors
Preface vii

Contents
Integrated Solutions in Computer-Based Control
Automatic System Based on Riesz MV-algebras, for Predictive
Maintenance of Bearings of Industrial Equipment Using Temperature
Sensors
.................................................. 3
Dan Noje, Radu Catalin Tarca, Nicolae Pop, Alin Octavian Moldovan,
and Ovidiu Gheorghe Moldovan
Smart Contracts: A Valuable Technology in ERP e-Commerce
Systems and for Customer Experience
.......................... 20
Maria Georgiana Stoica
A Nursing Mobile Robot for COVID19 Patients Using a RPI
and Mobile Device
......................................... 34
Víctor Lomas-Barrie, Mario Peña-Cabrera, Mauricio Reyes-Castillo,
T. Alcantara-Concepcion, and Gastón Lefranc
Electric Vehicles Between Recycling and Sustainable
Development - @.ro
........................................ 47
Razvan Marcus, Cosmin Rus, Monica Leba, and Marius Risteiu
Conversion of a Hybrid Flexible Manufacturing Cell
to Industry 4.0
............................................ 63
Valente Váquez-Velazquez, Mario Peña-Cabrera, Víctor Lomas-Barrié,
and Gastón Lefranc
A New Way of Collecting Trafﬁc Data Inside a Smart City Based
on IoT Devices and Communication Protocols Designed
for Preserving Perfect Anonymity
............................. 72
Nicolaie Popescu-Bodorin and Mariana Nagy
What UAE Software Students Think About Software Testing:
A Replicated Study
......................................... 85
Luiz Fernando Capretz, Saad Harous, and Ali Bou Nassif
ix

The Impact of COVID-19 on the Air Transportation Network.......94
Sergey Shvydun
Advanced Control Systems Integrating Computers and
Communications
Linear and Nonlinear Observers Developed for Direct Current
Electric Drive Systems
...................................... 107
Alexandra-Iulia Szedlak-Stinean, Radu-Emil Precup,
and Raul-Cristian Roman
Energy-Efﬁcient Wireless Sensor Networks for Greenhouse
Management
.............................................. 120
Attila Simo and Simona Dzitac
Digitization of the Characteristic Parameters of a Greenhouse
in Order to Streamline Energy Consumption
..................... 135
Simona Dzitac, Adrian Cheregi, Attila Simo, and Gabriela Elena Badea
Planning the Electricity Transactions Based on AI
................ 144
Victorita Radulescu and Bogdan Alexandru Radulescu
Integrated Solutions and Computerized Human Gesture Control
.....156
Bogdan Alexandru Radulescu
Research on the Storage of Integrated Pipe Corridor Monitoring Data
Based on Blockchain Technology
.............................. 167
Chunxiang Sun, Dan Chang, Daqing Gong, and Mincong Tang
Review of Trends in Manufacturing Systems Based
on Industry 4.0: The Opportunities
............................ 182
Gastón Lefranc
Soft Computing Including Fuzzy System Approach
Non-parametric Rank Correlation Trained Single-Hidden Layer
Feedforward Neural Networks for Medical Data
.................. 195
Smaranda Belciug and Renato Constantin Ivanescu
Nonstandard Fuzzy Sets: A General View
....................... 208
Sorin Nădăban and Dan Deac
Extension-Principle-Based Approach to Least Square Fuzzy Linear
Regression
............................................... 219
Bogdana Stanojevićand Milan Stanojević
Fuzzy Logic Applied in Hierarchical Secondary Regulation
Voltage-Reactive Power
..................................... 229
Victorita Radulescu
x Contents

Modeling with Fuzzy Logic the Impact of Overhead Power Lines
on the Environment
........................................ 239
Victorita Radulescu
Fuzzy Fractal: An Information Entropy View
.................... 250
Qianli Zhou, Chenhui Qiang, and Yong Deng
IV-FEC: Information Volume-Based Fuzzy Evidential Classiﬁer
......260
Jixiang Deng and Yong Deng
Boosting Data Management Using Artiﬁcial Neural Networks
........270
Francisco Sepúlveda-Ortiz, Claudia Durán, Diego Fuentealba,
Manuel Vargas, and Raúl Carrasco
Decision Making and Support Systems
Algorithmic Modeling to Individualize Touristic Experience in Petrila
Theme Park
.............................................. 283
Ionela Samuil, Andreea Ionica, and Monica Leba
A Combined VIKOR-TOPSIS Method with Application in Cloud
Services Provider Selection
................................... 293
Constanta Zoie Radulescu, Marius Radulescu, and Radu Boncea
Two-Sample Permutation Test for Comparing Marginal Probabilities
of Multivariate Bernoulli Distributions
.......................... 303
Stefano Bonnini and Michela Borghesi
Measuring Consensus in Group Decision-Making Problems Through
an Inequality Measure
...................................... 313
J. M. Tapia, F. Chiclana, M. J. del Moral, and E. Herrera–Viedma
Some Trends in Fuzzy Decision Making
........................ 320
J. R. Trillo, F. J. Cabrerizo, F. Chiclana, M. A. Martínez,
and E. Herrera-Viedma
Short-Term Covid-19 Incidence Prediction in Countries Using
Clustering and Regression Analysis
............................ 333
Fuad Aleskerov, Sergey Demin, Alexey Myachin,
and Vyacheslav Yakuba
Importance of Flexibility in Design and Organizational Structure
.....343
Fabián Barrales, Claudia Dúran, Diego Fuentealba, and Raúl Carrasco
Decision Support Systems for Integrated Land Management and
Transport Infrastructure in Support of Climate Change Mitigation
and Pandemic
............................................. 353
Maria Vișan and Sorin Lenus Negrea
Contents xi

Research on Early-Warning of Financial Risks of Logistics
Enterprises
............................................... 366
Yi Song and Dan Chang
Trustworthy and Green Design
Bridging Reliability to Efﬁciency Consecutive Elegant and Simple
Design
.................................................. 387
Valeriu Beiu, Sorin-Horațiu Hoară, and Roxana-Mariana Beiu
On the Roots of Certain Reliability Polynomials
.................. 401
Leonard Dăuş, Vlad-Florin Drăgoi, Marilena Jianu, Dominic Bucerzan,
and Valeriu Beiu
Using Delta-Wye Transformations for Estimating Networks’
Reliability
................................................ 415
Marilena Jianu, Leonard Dăuş, Sorin-Horaţiu Hoară, and Valeriu Beiu
Extension of HyperAlgebra to SuperHyperAlgebra and Neutrosophic
SuperHyperAlgebra (Revisited)
............................... 427
Florentin Smarandache
Green AI from Kirchhoff to Shannon
.......................... 433
Mihai Tache, Sorin-Horațiu Hoară, Vlad-Florin Drăgoi, and Valeriu Beiu
Author Index
................................................ 445
xii Contents

Integrated Solutions in Computer-Based
Control

Automatic System Based on Riesz
MV-algebras, for Predictive Maintenance
of Bearings of Industrial Equipment
Using Temperature Sensors
Dan Noje
1(B)
, Radu Catalin Tarca
2
, Nicolae Pop
3
, Alin Octavian Moldovan
1
,
and Ovidiu Gheorghe Moldovan
4
1
Doctoral School of Industrial Engineering, University of Oradea,
Universit˘at
,ii, 1, 410087 Oradea, Romania
[email protected]
2
Department of Mechatronics, University of Oradea,
Universit˘at
,ii, 1, 410087 Oradea, Romania
[email protected]
3
Institute of Solid Mechanics of the Romanian Academy,
Sector 1, 15 Constantin Mille, 10141 Bucharest, Romania
4
Department of Mechatronics, University of Oradea,
Universit˘at
,ii, 1, 410087 Oradea, Romania
[email protected]
https://doctorat.uoradea.ro/en
Abstract.Predictive maintenance systems take a more and more
important place in increasing the reliability of industrial equipment, of
productivity and reducing production costs. There’s a constant focus
to develop new system architectures and in using the latest numerical
methods and algorithms in developing more and more reliable predic-
tive maintenance systems. Based on recent results in using Shepard local
approximation operators deﬁned in Riesz MV-Algebras for IoT devices
signals processing, in this article we propose a new architecture for a
predictive maintenance system by adding an extra layer (denoted Data
Validation Layer). This new layer is processing the acquired signals by
sensors such that to provide to the predictive algorithms a complete and
validated data set. The proposed architecture was validated by imple-
menting and testing a predictive maintenance system that is monitoring
the bearings of industrial equipment using temperature sensors.
Keywords:Predictive maintenance
·IoT·Industrial equipment·
Sensors·Riesz MV-Algebras
1 Introduction
Predictive Maintenance (PrM) isused throughout the industry for several reasons
mainly in order to reduce cost, minimize break-down times, improve the quality
of the process and also it can have a major impact in environment protection [29].
cﬀThe Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Dzitac et al. (Eds.): ICCCC 2022, AISC 1435, pp. 3–19, 2023.
https://doi.org/10.1007/978-3-031-16684-6
_1

4D.Nojeetal.
Many industries adopt PrM and rely on forecasting abilities of diﬀerent sys-
tem in order to ensure adequate safety and reliability of all systems [25]. In
today’s industry, smart products and components (equipped with embedded
smart devices) are wirelessly networked and remotely monitored in real-time
[8]. Maintenance strategies are responsible for the performance and the proper
operation of manufacturing systems, turning into increasingly more relevant for
the sustainability of manufacturing processes [14]. Recent developments in infor-
mation technology, data transmission and information processing, such as IoT
(Internet of Things) technology, allow the development of PrM systems to be
eﬃcient and costeﬀective. This makes new PrM techniques being adopted in
industries where this was either not technically possible or not cost-eﬀective
[25] . PrM tools are used to determine when speciﬁc maintenance is required.
The tool relies on continuous observation of the machine, characteristics of a
machine or process, and this permits maintenance to be performed only when
is needed [25]. The PrM [18,32] is an eﬀective tool in reducing operating costs
and the frequency interval for maintenance activities on industrial equipment
[26,28]. PrMcan be a key factor in the Industry 4.0 concept when the integration
of sensors and sensorial systems target the cost reduction of maintenance, the
implementation zero-waste production, and the reduction of down time caused
by failures [17,25]. In developing PrM strategies and tools, the IoT is taking
a more and more important role [1,24,30]. A signiﬁcant part of the literature
focuses on methods of acquiring sensor data and interpreting this data in order
to identify potential problems in the production process. In general, these tech-
niques are generally called conditionbased maintenance techniques for industrial
equipment and processes [15]. These techniques are divided in three categories,
each category uses signals from existing process sensors, uses signals from test
sensors or generates a test signal that is sent towards the equipment and the
response is evaluated [15].
Another important aspect of PrM and data acquisition is the why the data
is used and processed. Speciﬁcally if we look at the IoT based PrM systems,
machine learning (ML) models are employed to evaluate the acquired data and
for the decision making process [3]. The most advanced PrM techniques focuses
on using Artiﬁcial Intelligence (AI) techniques in combination with proven mea-
surement and data processing strategies [11] in order to obtain best results in
term of eﬃciency and cost eﬃciency. Accurate and reliable data measurement
and acquisition for PrM is the general goal [4]. In order to determine de accu-
racy and the results of data analysis using the AI models, generally Artiﬁcial
Neural Networks models are used [10,13]. This paper addresses the issue of data
integrity in PrM systems. As PrM becomes increasingly complex, PrM uses
more and more artiﬁcial intelligence algorithms [5] so, the need for the trans-
mitted data to be correct or to be compensated if some of them are missing
or damaged is becoming more and more important. There are many causes of
data integrity problems, such as data acquisition problems or signal transmission
errors caused by electromagnetic interference or other causes. Random errors can
also be a factor in distorting the information and integrity of the data which is

Automatic System Based on Riesz MV-algebras 5
then considered in the processing for the decision-making system of the PrM
component. If an IoT-based PrM system is considered, the importance of data
integrity is even greater. It is known that not all phenomena in the real world
can be described accurately, i.e., you cannot answer yes or no to all questions.
Consequently, classical logic cannot cope with these types of problems. The con-
cept of multivalent algebra, in short MV algebra, introduced by C.C. Chang
[6,7] allows the description of the inaccuracy associated with real phenomena.
MV-algebras are seen as algebraic correspondents of a multivalent logic, just as
Boolean algebras are the algebraic correspondents of classical logic. Based on
this algebraic structure, vectorial MV-algebras [21] were introduced to provide
instruments to develop approximation methods. Later, Riesz MV-algebras [9]
who are a particular type of vectorial MV-algebras were introduced and studied
[2]. Based on the Riesz MV-algebra structure of IoT signals [22,23] a new archi-
tecture of an automatic system for PrM of bearings of industrial equipment is
proposed and presented. The data used in this research is related to the bear-
ing housing temperature of an electric motor. Several research identiﬁed bearing
housings as critical points for data collection in PrM systems [10–12,16,31]. In
most application the drive system is a key point for PrMand the evolution of
bearing temperature (along with vibration) is generally monitored and used as
a reliable information regarding system health and performance.
2 Materials and Methods
As mentioned, the new proposed architecture of an automatic system for PrM is
based Riesz MV-Algebras of IoT signals, in fact is using Shepard local approx-
imation operators deﬁned on these Riesz MV-algebras [22,23].To validate the
new architecture was decided to implement and test an automatic system for
PrM of bearings of industrial equipment. Considering this, we have to map the
acquired signals by the temperature sensors with algebraic structures such that
to be able to use the proposed mathematical methods.
2.1 Autonomous Systems
2.2 Mathematic Fundamentals Used in IoT Signal Processing
In [23] was considered that the data sent by an IoT device on regular time
intervals is a time-based function
f[0,n]:→[0,2
t
−1] (1)
0 is considered the moment when the ﬁrst signal was sent, and that 1 is the length
of the time interval when a new signal is transmitted by an IoT device and t is
the number of bits used to store the values in computer memory. In other words,
in our case, the function f gives the bearing temperature at a speciﬁc moment.
In [19] as proved that the structure ([0,2
t
−1],⊕,λ,0) is an MV-algebra, if
the following deﬁnitions are used:

6D.Nojeetal.
x
λ
y= defmin(2
t
−1,x+y) (2)
λx=
def2
t
−1−x (3)
∀x, y∈[0,2
t
−1]
If is considered the external operation•:R
+×[0,2
t
−1]→[0,2
t
−1], deﬁned
as follows:
a•x=
defmin(2
t
−1,a·x). (4)
∀a∈R
+and f∀x∈[0,2
t
−1]
In [20] was proved that the structure ([0,2
t
−1],⊕,λ,0,•) is a vectorial MV-
algebra and it is easy to see that it is also a Riesz MV-algebra. In [23]the
Shepard local approximation operator was introduced as a function
S:[0,n]→A (5)
deﬁned as follows:
S(f,x)=⊕
xi∈B(x,r)
K
∈
|x−x i|
r
◦
∈
xi∈B(x,r)
K
∈
|x−x i|
r
◦•f(x
i) (6)
where⊕and•are the Riesz MV-algebra operations,
B(x, r)={y∈[0,n]||x−y|≤r}and a kernel, which is a strictly decreasing
functionK:[0,1]→R
+.
In the above deﬁnition the Shepard local approximation operator, xi are
the time intervals when new data was received, and x are the time intervals
when no data was received and that has to be approximated. For more accurate
monitoring process, we can consider that the data sent by an IoT device on
regular time intervals is a two variable function:
[0,n]×[0,m]→[0,2
t
−1] (7)
where the ﬁrst variable represents the time dependency as in the one variable
case, and the second variable is the bearing revolution speed. 0 for the sec-
ond variable is when the bearing is not moving at all, and m is considered as
maximum revolution speed. If we consider this approach, the deﬁnition of the
generalized Shepard local approximation operator should be used [22]. In this
case, the function f gives the bearing temperature at a speciﬁc moment and
dependent on a speciﬁc revolution speed.
2.3 Automatic System Based on Riesz MV-algebras, for PrM
of Bearings of Industrial Equipment Using Temperature Sensors
As it is described in [30] usually an IoT-based PrM system consists in ﬁve dif-
ferent components:

Automatic System Based on Riesz MV-algebras 7
– The Data Acquisition Layer consists in sensors placed in speciﬁc points of the
industrial equipment;
– The Data Communication Layer, is the layer that transfer the collected data
to a data storage;
– The Data Storage, which is the layer that store the data in structured way
such to be suitable for analysis processes and decision making;
– The Predictive Analysis Layer implements predictive algorithms, mostly
based on machine learning. They have to generate maintenance tasks in case
of a speciﬁc incident may occur;
– The Predictive Maintenance Software (PrMSw) is the layer that send notiﬁ-
cations to the maintenance team regarding the necessary maintenance tasks
that have to be performed and also provide all required information needed
for the task’s accomplishment. The PrMSw should also provide reports.
It is known that in the process of data acquisition may occur errors. Those
errors may be generated by noises that are altering the signal or by disconti-
nuity in data acquisition. Another known fact is that the predictive analysis
layer needs a complete and validated set of data in order to deliver accurate
predictions. Considering these facts, a new architecture of an IoT-based pre-
dictive maintenance system is proposed. A new layer to be added is proposed.
This new layer should be placed between the data acquisition layer and the data
communication layer, and we call it Data Validation Layer. The Data Validation
Layer has the role collect all the signals acquired by the sensors to ﬁlter the
altered values by noises and to ﬁll in the missing signals. This way, complete set
of validated signals is provided to the Data Communication Layer. The signal
processing algorithm used in the Data Validation Layer is the one presented in
Fig.1.
If we consider the bearing temperature as a time-based function, it is obvi-
ous that is a continuous function. Also, if we consider that the temperatures
are stored in memory, the possible values are in the range [0,2
t
−1], where t
represents the number of bits used to store the value. Considering the results
presented in [19], and using the operations as deﬁned in (1), (2) and (3), the
structure ([0,2
t
−1],⊕,λ,0,•) is a Riesz MV-algebra, means that the tempera-
ture values acquired by sensors from bearings has a Riesz MV-algebra structure.
Considering this, we can conclude that:
– If we consider that the bearing temperature is a time-based function, the
approximation algorithm introduced in [23] can be used. This algorithm is
based on the Shepard local approximation operator.
– In case that we consider that the bearing temperature is a time and bearing
revolution speed function, in the approximation algorithm, the Generalized
Shepard local approximation operator [22] can be used.
If we plan to optimize the transmission of data, the values collected in the
buﬀer will only be transmitted from time to time, depending on the available
connectivity. Also, the buﬀer should not be emptied because we still need some

8D.Nojeetal.
Fig. 1.The signal processing algorithm used in the Data Validation Layer
validated values for further computations. The number of values that should be
kept in the buﬀer depend on the approximation parametrization.
An interesting fact is that the approximation algorithm based on Shepard local
approximation operator can be used not only to generate the approximated
signal value, but also to determine if the signal is altered by noises.
The algorithm to determine if the signal is altered by noise, has the following
sequence:
1. We consider a signal value Sg;
2. A new signal value NSg is approximated based on previous stored values in
buﬀer;
3. If|Sg−NSg|≥Accepted error, the Signal value is considered altered by
noise.Machine learning or experiments can be used to set the Accepted error.

Automatic System Based on Riesz MV-algebras 9
3 Results
The experimental implementation of a predictive maintenance system as
described above, was implemented on an electric motor. The electric motor for
which the data acquisition on the temperature evolution was carried out is a
three phase motor with a power of 1.5 kW. The motor shaft is mounted in two
zones by means of two radial ball bearings. A bearing type 6206ZZ is mounted
on the front side and a bearing type 6205ZZ on the rear side. In order to moni-
tor the temperature development in the two bearing housings, two temperature
sensors are mounted for each bearing. The sensors used are LM 35 type sensors
produced by National Instruments. The sensors have a guaranteed accuracy (at
+25
◦
C) of 0.5
◦
C, the operating temperature range is−55
◦
to +150
◦
Cand
the operating voltage is between 4 and 30 V. The heat transfer was ensured by
using a thermally conductive paste applied between the motor housing and the
temperature sensor, as seen in Fig.2.
Fig. 2.The temperature sensors setup

10 D. Noje et al.
Fig. 3.The setup used for temperature acquisition.
Data acquisition from the four temperature sensors was done using a DATAQ
INSTRUMENTS acquisition board model DI-2018 and Windaq application
(Fig.3). The acquisition system used has a resolution of up to 16 bits and a
220 kHz sample throughput rate. To acquire data on the temperature evolution
in the bearings, 100 samples/second were acquired for each channel at a sam-
pling interval of 10 ms. The data acquisition time was 30 min. The acquisition
was made when the motor was under normal operating conditions, i.e., it was
forced ventilated by means of a fan mounted on the motor shaft (Fig.4).
Fig. 4.Acquired temperatures over 30 min period.
In the implementation was considered that the bearing temperature is a time-
based function and thus the approximation algorithm introduced [26]wasbe
used. The implemented algorithm (Algorithm 1) has approximated each missing
value based on the previously acquired temperatures that were in the proximity
of radius r of the missing value.In this case we have to compute each missing
value during the acquisition time. In this situation we can say that each time

Automatic System Based on Riesz MV-algebras 11
we encounter a missing value, the previous one exists, no matter if it is acquired
or approximated one. Considering these, we decided to use the parametrization
presented in [23], that provided the best results in case each second value is
missing, situation that correspond to our assumption that each value has at
least one acquired value before it.
The used parametrization is:
– the selected kernel was Exponential kernel:K(u)=e
−λu
2
–r=3and λ=10.
In the experiments were used the four sets of approximately 360000 values
of acquired temperatures by the four sensors.The acquired temperatures were
stored in a buﬀer and transferred to the Data Validation Layer once the buﬀer
is full. After the buﬀer content is transferred, the buﬀer is emptied, but the last
r values are kept, such that to be able to approximate the temperature in case
the ﬁrst value is missing after the buﬀer content is transferred.
The PrMSw used was PriDesk [27] a collaborative tool that ensure traceabil-
ity and process automation and a powerful reporting module. The integration
was realized using secure connection to its API. In order to reduce the complex-
ity of the implementation and since the focus of this research is to determine a
suitable implementation of the Data Validation Layer, we decided to create a
maintenance ticket in case the bearing temperature has exceeded a certain pre-
viously set threshold temperature. To prevent the ﬂood of the human operators
with unnecessary tickets, a new ticket is created not for each temperature that
exceeds the threshold temperature, but only in the situation that we switch from
a normal operating temperature to a temperature above the threshold one. Since
all the tests were very similar, for exempliﬁcation we present the results of the
tests performed on a set of 50 consecutive acquired temperatures with the help
of ﬁrst sensor.
We mention again that each every 10 ms a new temperature was acquired. In
order to simulate the case of a malfunction of the sensor, we randomly eliminate
some of the acquired temperatures and generate a new value instead. We con-
sidered two experiments. In the ﬁrst experiment, of which obtained results are
presented in Fig.5, we chose to consider that from time to time a single value
is missing. The approximated values are marked in Fig.5using a dark purple
point over the orange line.
In the second experiment, of which obtained results are presented in Fig.6,
we chose to consider that from time-to-time multiple values are missing. The
approximated values are marked in Fig.6using a dark purple point over the
orange line.
4 Discussion
Analysing the results of the performed experiments, we noticed that the approx-
imated values tend to keep the trend of previous acquired temperatures, thus
the approximation isn’t good enough. This led us to the idea to use another

12 D. Noje et al.
Fig. 5.Approximation of missing values using Algorithm 1 in experiment 1 for singular
missing values.
Fig. 6.Approximation of missing values using Algorithm 1 in experiment 2, for multiple
missing values.

Automatic System Based on Riesz MV-algebras 13
algorithm (Algorithm 2) that compute each missing value based on both pre-
vious and next acquired temperatures that are in the proximity of radius r of
the missing value. To implement this algorithm, we have to compute the missing
values just after the temperature buﬀer is full. In order to avoid the possibility
to be in the situation that the buﬀer starts or ends with missing values, in which
situation the approximation algorithm fails, in the Data Validation Layer, we
implemented a variable length buﬀer considering the following constrains:
1. If the new buﬀer that is created starts with missing values, the last r values
from the previous buﬀer will be kept as start in the new one. The r value is
the radius used to determine the values in the proximity of the missing one
that will be approximated;
2. If the new buﬀer that is created ends with missing values, it will be considered
as ending after the last subset of existing values. In this case, we try to keep
no more than r values from this subset end together with the missing values
as start of the next buﬀer. In order to avoid having this subset duplicated it
will be used only to approximate the missing values and will not be parsed
to the Data Storage Layer.
Considering the above premises, we have run the same two experiments that
have been used to test Algorithm 1:
– in the ﬁrst experiment, of which obtained results are presented in Fig.7,we
chose to consider that from time to time a single value is missing.
– in the second experiment, of which obtained results are presented in Fig.8,
we chose to consider that from time-to-time multiple values are missing.
Since we intended to compare the two algorithms in the same conditions, we
selected the same values as missing as for the Algorithm 1. In both Fig.7and
Fig.8, the approximated values computed using Algorithm 2 are marked using
a red point over the gray line.
As seen in the Fig.5and Fig.7, in most of the cases the Algorithm 2 gives us a
better result than Algorithm 1 in the case of missing singular values. This state-
ment is enforced by the maximum diﬀerence between the acquired temperature
and the approximated one:
– using Algorithm 1, the maximum diﬀerence was 0.06103;
– using Algorithm 2, the maximum diﬀerence was 0.01526.
Also, in the case of second experiment, when multiple values were consid-
ered missing, Algorithm 2 gave better results than Algorithm 1, as can be seen
in Fig.6and Fig.8. For this experiment the maximum diﬀerence between the
acquired temperature and the approximated one was:
– using Algorithm 1, the maximum diﬀerence was 0.09155;
– using Algorithm 2, the maximum diﬀerence was 0.04069.

14 D. Noje et al.
Fig. 7.Approximation of missing values using Algorithm 2 for singular missing values.
After the two experiments, we can conclude that Algorithm 2 is providing
better results no mater in which situation we are, thus is more suitable for our
purpose.
In [23], for multiple missing values, and using the Exponential Kernel, were
obtained better results using the following parametrization:r=5andλ= 10.
This led us to the idea to experiment this parametrization for our collected
temperatures too in the case of multiple missing values. For consistency, we
performed this experiment using both algorithms. If in the case of Algorithm
1, the result was not surprising, and with this parametrization it gave worst
results, as seen in Fig.9. In the case of Algorithm 2 not always we got better
results using this new parametrization, as seen in Fig.10.
In this case we noticed that there are two diﬀerent situations:
– when the temperature values are having a smoother variation. In this situation
the new para-metrization conﬁrms the results of [23];
– when the missing temperature is near a local minimum or local maximum,
because in our situa-tion the temperature variation is quite big next to these
points, the results of [23] are not conﬁrmed.
– in conclusion, we can state that the obtained results using Algorithm 2 are
better than Algorithm 1, thus we can conclude that Algorithm 2 is more suit-
able to be used to develop the Data Validation Layer. During the experiment,
new ideas raised to further develop the algorithms used by the Data Valida-
tion Layer and also to develop new IoT devices to detect in a more accurate
way if the bearing gives any signs that may start to perform abnormally. As
new research directions to further develop the newly introduced automatic

Automatic System Based on Riesz MV-algebras 15
Fig. 8.Approximation of missing values using Algorithm 2 for multiple missing values.
Fig. 9.Approximation of missing values using Algorithm 1 using parametrizationr=5
andλ= 10 for multiple missing values.
system based on Riesz MV-algebras for PrM of bearings of indus-trial equip-
ment using temperature sensors, we mention:
– considering the results of the two diﬀerent parametrizations used by the Algo-
rithm 2, we came to the conclusion that we could probably get better results
if we would develop a new algorithm that uses a mix of parametrizations for

16 D. Noje et al.
Fig. 10.Approximation of missing values using Algorithm 2 using parametrization
r= 5 andλ= 10 for multiple missing values
the radius r depending if we are or we are not in the region of a local minimum
or maximum;
– we have considered that a signal is altered by noise if the diﬀerence between
the acquired value and the approximated value is at least x
◦
C, but more
accurate detection algorithms can be used. Thus, a research idea would be
either to develop new algorithms to detect signals altered by noises or to
identify existing algorithms that are suitable for this purpose;
– to develop applications considering the bearing temperature as a time and
bearing revolution speed function. In this case we should use the generalized
approximation algorithm described in [22];
– to develop new IoT devices starting from other types of sensors, like vibration
or sound detection sensors.
5 Conclusions
This research started from the structure of an IoT-based predictive maintenance
system as it is described in [30] and from the necessity to provide a complete
set of validated data to the Predictive Analysis Layer such that all decisions
regarding the actions that have to be made to be as accurate as possible. In
order to achieve this goal, we proposed to add a supplementary layer, called
Data Validation Layer, in the structure of the analyzed IoT-based predictive
maintenance system. This supplementary layer roles to verify if the acquired
data by the IoT devices is altered by noises or is missing. If the data is altered by
noises, it is considered as missing. In both situations, a new value is generated

Automatic System Based on Riesz MV-algebras 17
using approximation algorithms and inserted in the acquired date stream. To
demonstrate the viability of this new architecture, we performed experiments
by eliminating some acquired temperatures of the bearings of an electric motor,
values that were considered either altered by noises or not acquired, with new
values generated using Shepard local approximation operator [23]. Were con-
sidered two implementations, the ﬁrst one is using for approximation only the
validated data acquired before the one that should be generated and a second
implementation that is using for approximation the validated data acquired both,
before and after the one that should be generated. The second implementation
has provided better results, but from our point of view both implementations
provided good enough results to validate the scope of the experiments. Also,
out of the performed experiments, we concluded that based on the possible
variation paten of the expected acquired values, several parametrizations of the
Algorithm 2 can be used. As further development of the Data Validation Layer,
can be considered:
– testing and/or developing new methods to detect the acquired data by sensors
that are aﬀected by noise;
– testing and/or developing new methods to approximate the acquired data by
sensors that were not validated (aﬀected by noise or missing);
– to optimize the data transfer, we can consider compressing the data contained
in the processing data buﬀer and after it is received by the Data Storage Layer
to be decompressed.
Funding Information.The research has been funded by the University of Oradea,
within the Grants Competition ”Scientiﬁc Research of Excellence Related to Priority
Areas with Capitalization through Technology Transfer: INO - TRANSFER - UO”,
Project No. 326/21.12.2021.
This project is funded by the Romanian Ministry of Research, Innovation and Digitisa-
tion through Programme 1 - Development of the National Research and Development
System, Subprogramme 1.2 - Institutional Performance - Projects for funding the excel-
lence in RDI, Contract No. 29 PFE / 30.12.2021 with University of Oradea.
References
1. Adila, A.S., Husam, A., Husi, G.: Towards the self-powered internet of things
(IoT) by energy harvesting: trends and technologies for green IoT. In: Proceedings
of the 2018 2nd International Symposium on Small-scale Intelligent Manufacturing
Systems (SIMS), pp. 1-5. IEEE, Cavan (2018)
2. Bede, B., Di Nola, A.: Elementary calculus in Riesz MV-algebras. Int. J. Approx.
Reason.36, 129–149 (2004)
3. Cakir, M., Guvenc, M.A., Mistikoglu, S.: The experimental application of pop-
ular machine learning algorithms on predictive maintenance and the design of
IIoT based condition monitoring system. Comput. Ind. Eng.151, 106948 (2021).
https://doi.org/10.1016/j.cie.2020.106948

18 D. Noje et al.
4. Cardona, M., Cifuentes, M., Hernandez, B., Prado, W.: A case study on remote
instrumentation of vibration and temperature in bearing housings. JLPEA11,44
(2021).https://doi.org/10.3390/jlpea11040044
5. Carvalho, T.P., Soares, F.A.A.M.N., Vita, R., Francisco, R.D.P., Basto, J.P.,
Alcal´a, S.G.S.: A systematic literature review of machine learning methods applied
to predictive maintenance. Comput. Ind. Eng.137, 106024 (2019).https://doi.org/
10.1016/j.cie.2019.106024
6. Chang, C.C.: Algebraic analysis of many valued logics. Trans. Amer. Math. Soc.
88, 467–490 (1958)
7. Chang, C.C.: A new proof of the completeness of the Lukasiewicz axioms. Trans.
Amer. Math. Soc.93, 74–80 (1959)
8. Csokmai, L.S., T¸arc˘a, R.C., Bung˘au, C., Husi, G.A.: Comprehensive approach to
oﬀ-line advanced error troubleshooting in intelligent manufacturing systems. Int.
J. Comput. Commun.2014(10), 30 (2014).https://doi.org/10.15837/ijccc.2015.1.
1561
9. Di Nola, A., Flondor, P., Leustean, I.: 2004. MV-Modules J. Algebra261, 21–40
(2003)
10. Esim, E., Yıldırım, S¸: Drilling performance analysis of drill column machine using
proposed neural networks. Neural Comput. Appl.28(1), 79–90 (2016).https://
doi.org/10.1007/s00521-016-2322-8
11. Falekas, G., Karlis, A.: Digital twin in electrical machine control and predictive
maintenance: state of the art and future prospects. Energies14, 5933 (2021).
https://doi.org/10.3390/en14185933
12. Faysal, A., Ngui, W.K., Lim, M.H., Leong, M.S.: Noise eliminated ensemble empir-
ical mode decomposition scalogram analysis for rotating machinery fault diagnosis.
Sensors21, 8114 (2021).https://doi.org/10.3390/s21238114
13. Fontes Godoy, W., Morinigo-Sotelo, D., Duque-Perez, O., Nunes da Silva, I., Goed-
tel, A., Pal´acios, R.H.C.: Estimation of bearing fault severity in line connected and
inverter fed three phase induction motors Energies,13, 3481 (2020).https://doi.
org/10.3390/en13133481
14. Franciosi, C., Voisin, A., Miranda, S., Iung, B.: Integration of I4.0 technologies
with maintenance processes: what are the eﬀects on sustainable manufacturing?
IFAC-PapersOnLine,53, 1-6 (2020).https://doi.org/10.1016/j.ifacol.2020.11.001
15. Hashemian, H.M.: State-of-the-art predictive maintenance techniques. IEEE Trans.
Instrum. Meas.60, 226–236 (2011).https://doi.org/10.1109/TIM.2010.2047662
16. Hou, Z., Lv, X., Zhuang, S.: Optimized extreme learning machine-based main bear-
ing temperature monitoring considering ambient conditions’ eﬀects. Energies14,
7529 (2021).https://doi.org/10.3390/en14227529
17. Li, D., Landstr¨om, A., Fast-Berglund,˚A., Almstr¨om, P.: Human-centred dissem-
ination of data information and knowledge in industry 4.0. Procedia CIRP84,
380–386 (2019).https://doi.org/10.1016/j.procir.2019.04.261
18. Mobley, R.K.: An Introduction to Predictive Maintenance, 2nd edn. Butterworth-
Heinemann, Amsterdam, New York (2002), ISBN 978-0-7506-7531-4
19. Noje, D., Bede, B.: The MV-algebra structure of RGB model. Stud. Univ. Babes-
Bolyai, Informatica56, 77–86 (2001)
20. Noje, D.: Using Bernstein polynomials for image zooming. In: Proceedings of
the Symposium Zilele Academice Clujene, Computer Science Section, pp. 99–102
(2002)
21. Noje, D., Bede, B.: Vectorial MV-algebras. Soft Comput.7(4), 258–262 (2003)

Automatic System Based on Riesz MV-algebras 19
22. Noje, D., Tarca, R., Dzitac, I., Pop, N.: IoT devices signals processing based on
multi-dimensional shepard local approximation operators in Riesz MV-algebras.
Int. J. Comput. Commun.14, 56–62.https://doi.org/10.15837/ijccc.2019.1.3490
23. Noje, D., Dzitac, I., Pop, N., Tarca, R.: IoT devices signals processing based on
Shepard local approximation operators deﬁned in Riesz MV-algebras. Informatica
31(1), 131–142 (2020).https://doi.org/10.15388/20-INFOR395
24. Parpala, R.C., Iacob, R.: Application of IoT concept on predictive maintenance
of industrial equipment. In: MATEC Web Conference, vol. 121, p. 02008 (2017).
https://doi.org/10.1051/matecconf/201712102008
25. Pech, M., Vrchota, J., Bedn´aˇr, J.: Predictive maintenance and intelligent sensors
in smart factory: review. Sensors021(21), 1470 (2021).https://doi.org/10.3390/
s21041470
26.https://www.avnet.com/wps/portal/abacus/solutions/markets/industrial/
predictive-maintenance-iot/
27.https://pri-desk.ro/
28.https://www.scnsoft.com/blog/iot-predictive-maintenance-guide
29. Selcuk, S.: Predictive maintenance, its implementation and latest trends. Proc.
Inst. Mech. Eng. Part B: J. Eng. Manuf.231, 1670–1679 (2017).https://doi.org/
10.1177/0954405415601640
30.https://www.softwareadvice.com/resources/iot-predictive-maintenance/
31. van den Hoogen, J., Bloemheuvel, S., Atzmueller, M.: Classifying multivariate sig-
nals in rolling bearing fault detection using adaptive wide-kernel CNNs. Appl. Sci.
11, 11429 (2021).https://doi.org/10.3390/app112311429
32. Zerhouni, N.: From Prognostics and Health Systems Management to Predictive
Maintenance 1: Monitoring and Prognostics. ISTE Ltd/ Wiley Hoboken, NJ (2016)

Smart Contracts: A Valuable Technology
in ERP e-Commerce Systems and for Customer
Experience
Maria Georgiana Stoica
(B)
Department of Economic Informatics and Cybernetics, Bucharest University of Economic
Studies, Romana Square 6, 010374 Bucharest, Romania
[email protected]
Abstract.In the past few years, more and more e-commerce companies are pro-
gressively looking for business opportunities and ways to evolve in terms of their
informatic system in such a crowded market and gain a unique customer expe-
rience to earn the client’s trust. This need is exacerbated, taking into account
the competition generated by the major market leaders, by constantly introduc-
ing innovations in planning the business resources, and by the results-oriented for
stimulating the competition. This study theoretically systematizes how blockchain
technology and smart contracts are slowly becoming a necessity in the ERP sys-
tems from the big companies that offer online shopping services, whether B2B
or B2C, providing extra security for its clients. The models deﬁned in this study
highlight the technical efﬁciency in a resource system planning of three compa-
nies, market leaders in Romania, in terms of supply-chain, secure payments, and
forecast, for comparing the costs of the change when using blockchain technol-
ogy. Moreover, for a better shopping experience in online stores, the efﬁciency
and effectiveness scores of applying these technologies to the customer review
section, so that frauds are analyzed and can be signiﬁcantly reduced by making in
this way a secured and more reliable system. The study analysis of these aspects
shows that the efﬁciency of minimizing frauds and improving the technology of
ERP development by creating a safe and stable environment varies substantially
between models and leads to a new era in e-commerce that is so needed nowadays.
Keywords:Blockchain∙e-commerce∙ERP
1 Introduction
With the rise of globalization and the perpetual change of individual’s lifestyle according
to nowadays needs, in the e-commerce domain, there is a growing demand to cover all
the customer needs, but also an evolution towards a technology that gives business proﬁt
and minimization of business costs. A decisive factor in the signiﬁcant emphasis of
the business performance analysis in the e-commerce business is the current pandemic,
which has risen in the last 3 years the need to increase IT and ERP systems, due to the
huge demand of orders and online transactions. Due to this, large volumes of customers
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Dzitac et al. (Eds.): ICCCC 2022, AISC 1435, pp. 20–33, 2023.
https://doi.org/10.1007/978-3-031-16684-6
_2

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 21
are directed to the major market leaders, because as a buyer, the decision is to choose
the most stable, secure, and reliable IT product, through which he can meet his need.
From the company’s point of view, they are facing the satisfaction and maintenance of
the shop characteristics that customer is attracted, but also the fact that a high evolution
is needed to keep competitors at distance and to face with technology which continues
to evolve and surprise through its new applications. Given these aspects, the blockchain
is considered a remarkable technological step forward that leaves a signiﬁcant touch on
everyday business, especially in the ﬁeld of e-commerce.
According to the World Economic Forum [11], this technology has the potential
to become a powerful tool for transactions, so it helps users gain trust, boost business
and drastically reduce corruption. More speciﬁcally, if we take as an example the big
market leaders in e-commerce such as Amazon, Carrefour, or IBM, they have already
approached blockchain technology since 2019 in their solutions for the supply chain,
but also for verifying the authenticity of customer goods. More speciﬁcally, through
the blockchain technology applied to SCM, companies can document and keep track
of production updates in a single shared register, which has the role of providing trans-
parency to the data, and more importantly, the data source will always be correct. In this
way, from the stage of production to the stage of delivery of the product to the client,
companies can query the status and location of the ordered product at any time. This
helps to combat issues such as counterfeit goods, breaches of compliance, delays, and
waste [9]. In contrast to these international models, this technology has not yet been
adopted nationally in Romania, in the ﬁeld of e-commerce.
Having all this information about the business-level technological models of the
big e-commerce giants, the current study simulates and make a detailed analysis of the
integration of blockchain technology and smart contracts on the SCM, payments, and
statistical forecasts at the level of company costs, but also it will be explaining a method to
avoid fraud at the level of client reviews. The study will detail two perspectives, namely:
the company and the client. Also, for an accuracy of the data and the evolution of this
study, we will analyze the aspects of 3 companies from Romania, eMAG, Fashion Days,
and Altex, and we will analyze the growths and changes that will take place on users as
well as businesses in case integration of blockchain technology and smart contracts at
ERP level in e-commerce companies.
2 Background
I will start by reviewing the existing literature and research on technologies applied to
e-commerce systems and how the strategy of the resource planning system can inﬂuence
the value of system security and customer conﬁdence. Then I will identify the reasons
why a customer is inﬂuenced to shop online and the issues that need to be improved
technologically. I will consider 3 online stores in Romania, namely eMAG, Fashion
Days, and Altex and I will follow the aspects that I speciﬁed in correlation with the
international models that have adopted smart contracts at the ERP level to minimize
business costs and client conﬁdence. I will refer to the Amazon e-store as a reference
for its successful integration of blockchain technology.
Several studies are researching the impact of smart contracts and how blockchain
technology inﬂuences ERP systems or more, the supply chains. In a study, it is making

22 M. G. Stoica
detailed references on how to achieve real-time inventory visibility, full visibility, and
real-time updates for the price, product catalog, price breaks, and many others aspects.
In comparison with this study, it does not offer a clear architecture or design of such a
service, but theoretically presented aspects regarding the automation of many expensive
processes and integrations that can be done in real-time, such as Demand and Supply
Management, Logistics, and Manufacturing [16]. An important thing in this study is the
advantages offered for stakeholders in a business that integrates smart contracts, but also
how the origin of the products can be found, and more importantly the state, the origin
of the ingredients or materials that make up the product. The authors also propose to
record this data through a source map.
Another notable journal study provides general information on the implications of
blockchain in industry 4 and details the supply chain, but for the ﬁeld of construction.
This time, it is presented an architectural model of creating a database that manages the
supply chain and uses blockchain. By comparison to this study, it does not detail the entire
level of ERP, but only details the process of data management and the beneﬁts of their
transparency [17]. Also, in another article in the International Journal of Information
Management, the study assesses proﬁtability geographically in India but does not cover
all aspects of an SCM, such as supplier, buyer performance, integration times, and
technology cost minimization [18]. However, the demographic proﬁt, Reliability, and
AVE values brought by the blockchain integration in the SCM are summarized.
A more practical study, which was applied to a range of 236 people, which, although
not in the ﬁeld of eCommerce, presents concrete and measurable information on the
proﬁtability of the blockchain brought at the level of SCM, improving in 6 dimensions
this aspect. It is also concluded to improve the reliability of delivery and mass customiza-
tion which signiﬁcantly increases the proﬁtability of companies [19]. The dimensions
analyzed would be CRM, information quality, company improvement, product quality,
SCM, and mass customization. Interestingly, it was concluded that among many other
good practices in using blockchain, one would be integrated into it when the main players
in the supply chain are mistrust [19].
2.1 Empirical Regularities on National and International e-Commerce Trends
Empirical research at the national level, more precisely in Romania, results that any
improvement in e-commerce technologies inevitably leads to a sharp increase in
employee productivity, in this way facilitating costly processes [13]. At the same time,
there is a progression in the efﬁciency of the IT system, which leads to an increase in
the competitiveness of companies and the acceleration of interdependencies between
national economies and markets. For this reason, how e-commerce is constructed is a
decisive factor in the activity of economic organizations, so it leads to better management
of forecasting strategies and estimation of decisions taken at the level of the planning
system [2].
The 3 online stores, eMAG, Fashion Days, and Altex, leaders on the Romanian
national market, have similar technologies, but none of them use blockchain in the
resource planning system. EMAG and Fashion Days employ ERPs closely related to
a Business Intelligence solution. The ﬁrst two are part of the same company, so they
use similar technologies in their approach to resource planning systems. To maintain

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 23
its market-leading position in sales, the ﬁrst two e-stores have chosen Hostway as their
hosting provider and now operate on more than 40 IBM and Dell servers. Also, in
addition to Altex, the latter do both B2B and B2C trade, so they offer other companies
services through which they can sell and supply their products through eMAG. In terms of
customer satisfaction for the supply chain, Altex manages to deliver the ordered products
with an average greater than eMAG. This is due to the technologies approached in the
supply chain management system so that any improvement in the technology results in
faster delivery of orders. There is also no specialized track and trace system, especially
at the Altex e-shop. This has, as a consequence for the customer, the dissatisfaction with
the need to track the goods, but also an inefﬁcient forecast for the company because it is
not possible to accurately track the duration and route of the package to the destination.
To avoid fraud at a high level in the supply chain, blockchain technology is a great
success.
Analyzing the mentioned e-commerce websites, both eMAG, Altex, and Fashion-
Days, use as payment method PayU, which is one of the oldest electronic payment
processors since 2006 in Romania. After completing an order, a request is made to the
server, where an API is called from inside the ERP to PayU, where the payment will be
processed, and it will return with an answer to the order completion page.
Most e-stores in Romania use robust payment gateway solutions and services. Similar
is the payment gateway service provider, PayU, because it simpliﬁes the connection
between the customer and the business by making payments and ensuring that sensitive
information is kept secure and nothing else more. The question is how this guarantee can
be improved and how e-commerce can evolve into future technology, by transforming
the business into the future, and the answer is blockchain.
Why blockchain? Even though the existing traditional payment systems are still
effective, they have major limitations, such as delays in cross-border transactions, lack
of process automation, and involvement of intermediaries. Moreover, there is a growing
need to improve payment systems in areas such as the supply chain and retail, where, of
course, it is essential to maintain a transparent balance between the ﬂow of goods and
money. Blockchain-based payments offered by Gateway services, such as Mastercard
Blockchain, Samsung Blockchain Platform, Solana Pay, and many more, have become
increasingly popular due to the promising future that it brings, in which digital currencies
predominate and money digital moves through the internet as data, which means without
censorship and intermediaries charging each transaction. These services are known for
their speed and low transaction costs, and Solana is one of the best performing digital
currencies in recent years, with has great potential when it comes to solving some of
Ethereum’s most pressing issues [14]. Blockchain-based payments get so much faster
transaction processing times and offer consumers payment with a wide range of digital
currencies from their e-wallet directly to the merchant’s account, immediately settling
the measured costs in cash fractions. The Solana Pay payment gateway guarantees the
processing of over 50,000 transactions per second [14] which is amazing compared to
about 100 payments per second as offered by a regular payment API, which means
that Blockchain solutions are 500 times faster than traditional ones. This incorporates
consumer loyalty but also the conﬁdence of the integration of this technology by business

24 M. G. Stoica
people in their companies and thus becomes real weapons for the future of e-commerce
platforms and businesses.
What companies want is to secure their data, to effectively collect customer data, such
as name, age, buyer proﬁle, and this can be strengthened and improved by integrating
blockchain solutions into e-commerce systems to provide complete security and wallet
protection. A blockchain-based currency requires consumers to disclose sensitive data.
Moreover, considering a blockchain encryption algorithm that does not fail, the obvi-
ous conclusion is that by implementing this technology of the future, fraud is minimized,
and a secure, stable, and fast ERP is created, which offers security and customer con-
ﬁdence and certainly, the conversion rate, from the visitor status of the e-commerce
platform to buyer status, will be higher than 3%, as estimated on average [15].
Conversion Rate=Number of Customers/Number of Visitors;
Therefore, taking into account the payment method, Blockchain technology offers
more secure and credible solutions and services for e-commerce businesses, combining
transparency in transactions with more reliable management of consumer data. After all,
the beneﬁts of blockchain for e-commerce extend to cheaper business processes, greater
data security, and a substantially improved customer experience.
In contrast to Romania’s national models, Amazon has reinvented e-commerce sup-
ply chain management by implementing smart contracts in SCM. The company focuses
on high response capacity, combined with warehouse automation and robotics, which
results in one-day order delivery and even offers two-hour delivery with Prime Now
technology. Due to this, Amazon is attracting more and more customers from various
geographical areas and gaining their trust by massively fulﬁlling orders. In September
2019, Amazon purchased 100,000 electric delivery vans, and in 2020, the company took
advantage of low prices during the coronavirus pandemic to expand its means and vehi-
cles for air transportation. As estimates for the future, Amazon aims to deliver drone
packages to customers with Amazon Prime Air, to minimize the length of the goods
delivery process [12].
In this way, Amazon is becoming an international e-commerce ideal, thanks to its
well-developed ERP system, but also for the innovation of acceptance and integration
of new technologies, such as the implementation of smart contracts in the SCM system
and beyond. Romania is lagging slowly towards this business ideal, but there is still a
lot of work to be done to reach this IT potential, as the current technology we have thus
limited the resource planning systems as well as the satisfaction index of customers,
keeping on a steady trend.
2.2 Correlation Between SCM and Customer Satisfaction on Regular ERP
Systems
In the digital economy, the study of buyers focuses primarily on technologies that can
change the demand curve and, secondly, on interaction with buyers to stimulate demand
in their favor. Therefore, I will further present through a predictive analysis performed
between the independent variable that often inﬂuences the customer towards the purchase
of a product, the quality, and the dependent variable the customer’s expectations for the
ordered product.

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 25
To emphasize some measures for the client expectations I analyzed an existing dataset
from xlstat.com [8], which is a web platform that collects data. The results are based on
real-life data, where 250 customers of mobile phone operators have been asked several
questions to be able to model their loyalty. Each latent variable is related to one or more
manifest variables that are measured [8]. In this case, the manifest variables questions
are on a 0–100 scale. For example, for the Satisfaction latent variable, the ﬁve manifest
variables are: the product can be trusted in what it says and does, it is stable and functional,
it is delivered fast, the quality is good, the reviews are not fake.
It can be seen in Table1that the independent variable, namely the quality of the
products, has the most important effect on customer satisfaction, followed by the per-
ceived value. The impact of expectations is insigniﬁcant. At the same time, it is observed
that the independent variable given by the quality perceived at the client’s level in an
e-commerce system has a contribution of approximately 64.8% to R
2
.
Table 1.Equation of the model: Quality=0.556741866775216 * Expectation
ANOVA
Latent
variable
ValueStandard error t Pr>|t|f
2
Critical
ratio
The lower
bound
(95%)
Upper bound
(95%)
Intercept0.0000.053 0.0001.000 −0.5800.0000.000
Expectation0.5570.053 10.5550.0000.44910.7230.4330.648
As such, approximately 35.2% of the variation in customer satisfaction is explained
by other external factors, which may be the payment methods, the trust in the respective
electronic store, the inﬂuence of the reviews. Therefore, according to Fig.1,itwillbe
analyzed to what extent the change in technology brought to an IT system related to the
e-store, can inﬂuence the value of 35.2% of foreign factors and how we can increase
buyers’ conﬁdence and ERP performance in that company.
3 An Empirical Strategy for Building Client’s Trust
by Implementing Smart Contracts
I will start by presenting and theorizing the ideal model of an ERP system based on
building smart contracts on the blockchain and in what proportion it is more signiﬁcant
than the traditional model. Also, a comparative analysis was made between the proposed
system and the existing ones, that will illustrate the beneﬁts of the blockchain solution
and is being considered the impact of its implementation on the supply chain, payments,
and how signiﬁcantly improves economic forecasting.
3.1 A Theoretical Blockchain Model for ERP System Beneﬁts
From managing payments and reviewing audits to tracking inventory and assets,
blockchain technology allows for a much larger supply chain platform than ever before.

26 M. G. Stoica
At the same time, the electronic payment model, which would be an intermediate step in
the SCM process, is an important step that needs to be changed by implementing smart
contracts, to increase customer conﬁdence, but also to reduce costs for the currently
used model nationwide. The companies mentioned in the previous chapter, namely 3D
Secure technology [1].
Regardless of the intended application, blockchain technology offers so many advan-
tages to the current system that Romania also tends to adapt to this already existing
technology at the international level in the ﬁeld of e-commerce.
Some of the most important advantages over the regular supply chain system are
increased transparency, greater scalability, better security, and increased overall innova-
tion. The biggest advantage of the blockchain is its transparency. By documenting the
process of migrating a product along the entire supply chain, manufacturers can thus
reduce withdrawals by sharing logs with regulators and all relevant parties in the event
of a problem [5]. In this sense, it will increase customer conﬁdence, but also better man-
agement at the company level, and last but not least, a powerful tool in building future
statistical and economic forecasts at the data level.
Table 2.Comparison between blockchain and traditional SCM in eCommerce [7]
Comparative item Blockchain-based SCM Traditional SCM
Data integrity Tamper proof No temper proof
Data security Decentralized protection Centralized protection
Data veriﬁcation Archived Not archived
Government regulationsUsing authentication protocolNo auth protocol
I have summarized above in Table2a comparative detailed analysis about the beneﬁts
of Blockchain for a traditional e-commerce system in the supply chain management
sector.
Blockchain can provide greater security than current strategies. Virtually any number
of participants can access information from any number of contact points, providing ﬂex-
ibility as well as security throughout the system. Security is an advantage of blockchain
technology, as a permanently shared registry with hard-coded rules could reduce or
eliminate the need for audits required by internal systems and business processes [7].
Finally, enhanced innovation is a huge advantage over our current technology, with
new and specialized uses being able to be created from Blockchain technology to provide
an extra layer of encrypted protection, which tends to be an easily scalable and transparent
architecture.
3.2 Proposed Model for e-Commerce ERP System
In this section, I am proposing a model, based on blockchain and distributed systems,
attached to any e-commerce company for supply chain management and the visibility
of payments in the system.

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 27
Fig. 1.A supply chain management blockchain architecture
Figure1depicts at the architectural level the links that are created in supply chain
management by applying blockchain technology. Basically, at the SCM level, the visi-
bility increases at the level of each step of the product migration from the manufacturer,
to the seller and then to the customer. During this time a thorough check is made and
each recorded process is stored from the beginning of each event until it transits in each
node representative of the SCM process. The payment process is visible and immutable,
being added to the blockchain, and this aspect favors the minimization of frauds that can
be made at the level of payments between producer and seller [6].
In an e-commerce system that conducts either a B2B or B2C business, the relationship
between the company and the supplier is as important as the business and customer
relationship. Thus, the architecture presented above strengthens the link between the
manufacturer and the seller, and also between the e-commerce business and the producers
from whom the supply is made. This is possible through the visibility and transparency
of the data provided to the mentioned actors, thus gaining performance on the trust
between them.
Next, I analyze the activity at the blockchain level, a process visible in Fig.1, above,
and I will explain the beneﬁts of each process built in the proposed algorithm for supply
and order management.
In this way, Fig.2diagrams the exchange of messages at the level of micro-
services between the service of Manufacturer, Transport, Transport points, Delivery,
and Customer service. Speciﬁcally, each micro-service communicates with the other
micro-services by exchanging messages.
At the ERP level, the micro-services are intended to be implemented through the
Apache Kafka service-level architecture as is shown in Fig.3below. Single Kafka clus-
ters can even be deployed over multiple regions and globally. Mission-critical deploy-
ments without downtime or data loss are crucial for Blockchain and many other use
cases. Architecture patterns for the distributed, hybrid, edge and global, Apache Kafka

28 M. G. Stoica
Fig. 2.Activity diagram for blockchain process in SCM
deployments explore these topics in detail. Kafka will be run as a cluster of one or
more micro-services, which can span multiple data centers or cloud regions, such as the
Client-side or the Consolidated point region. They continuously import and export data
through event streams, making it possible to integrate this architecture with blockchain.
Fig. 3.Transaction’s component diagram for integration Kafka with blockchain
Also, for the infrastructure, a big step is to use or build a Kafka-native blockchain
infrastructure, as in Fig.3, so the deployment will be sequentially by the topics and the
blocks in the consumer and producer Kafka API protocol. It tends to get high scalabil-
ity and high-volume data processing in real-time for mission-critical deployments is a
crucial differentiator of using Kafka compared to traditional blockchain deployments.
Kafka enables a blockchain for real-time processing and analysis of historical data with

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 29
one single platform. Messages are alternatively encrypted and hashed consecutively.
The library methods are called inside a Kafka application’s message producer code to
wrap messages and called inside the application’s customer code to unwrap messages.
Since blockchains must be entirely successively requested, Kafka blockchain points
must either have a single partition, or consumers for each segment must participate to
arrange and sequence the records.
At the same time, the integration of the two technologies signiﬁcantly minimizes
data loss, because if one of the Kafka servers loses the connection or no longer works
in normal parameters, the other servers will take over the work to ensure continuous
operations without any data loss.
Therefore, it will greatly increase the visibility and transparency of data in supply
chain processes and payments made between the main actors of the SCM process, having
as indisputable effects the increase of technical performance as well as the improvement
of operational processes.
3.3 Blockchain as a Fake Reviews Solution
In addition to the beneﬁts presented above, blockchain also has the power in e-commerce
business to conﬁrm whether the reviews brought by its supposed customers are true or
fake. With the ability to create a digital map of the customer, which cannot be modiﬁed,
the blockchain can thus check whether or not users who have left a review have purchased
on the site before publishing them. More speciﬁcally, when buying something new or
trying out a new service, many of us head online to check what other customers think.
Over 80% of adults in the United States occasionally read online customer reviews before
buying something for the ﬁrst time. But only half of those who read reviews think they
give an accurate description of the product, according to the Pew Research Center [10].
So, the blockchain has the power to track customer information such as if he ordered
before and if the review is about what he ordered, and if it is relevant.
Customer reviews are especially important in an e-commerce business because they
have a signiﬁcant inﬂuence when a customer chooses whether or not to buy a product.
Beyond these beneﬁts, blockchain technology, used in conjunction with machine learn-
ing, can match a review with a real and appropriate user. In this case, the services can
signiﬁcantly reduce fraud and allow people to convey their opinion about their purchases,
the quality of orders, the duration of transport, or the quality of the products purchased
[4].
The system runs on a blockchain with powerful algorithms that take into account
customer preferences for features such as location, language, product or brand pref-
erences, cost, and other useful factors to provide real, useful, and appropriate product
reviews. Which customers follow. False reviews can be prevented using a veriﬁed dig-
ital identiﬁcation, supported by blockchain immutability and security. By combining a
biometrically authenticated ID with a payment source, you can create an unquestionable
way to authenticate reviews. An example is implemented by Amazon which consists of
biometric veriﬁcation of your digital identity when buying a product. One day after you
have purchased with the funded decentralized identity, the courier delivers the package
to you, by pinning your digital ID to allow contactless delivery. After this, the customer
leaves a sincere review of your digital identity veriﬁed by payment [12].

30 M. G. Stoica
Fig. 4.The logic schema for blockchain review algorithm
Figure4illustrates this, and so the algorithm is proposed by combining blockchain
technology with machine learning to increase data accuracy so that a real proﬁle of each
client will be created by ML and veriﬁed by blockchain whether or not it is a client.
4 The Effects of Smart Contracts and ERP Performance
Evaluations
Blockchain gives companies the ability to have different types of information in their
resource planning system, which also gives the parties easy access to information. For
example, in the traditional supply chain process, the transit of products until they are
delivered to customers is not easily monitored. Blockchain integration would provide
real-time product traceability.
Once a smart contract is integrated into an organization’s ERP system, it optimizes
operations, internal data control, and business processes, such as business-to-business
transactions. For example, as discussed in previous chapters, smart contracts can help to
automatically verify a transaction between companies, speciﬁcally in the SCM between
the manufacturer and the seller [6]. When integrating the blockchain into an organi-
zation’s ERP system, this should ideally be done with minimal disruption to normal
business processes and avoid hampering the processing and veriﬁcation of data and
event algorithms.
Blockchain within the supply chain signiﬁcantly progresses straightforwardness,
quality, security, and handle keenness through the secure record. Essentially, within the
fabricating segment, Blockchain empowers unwavering quality and works as a trust-
based layer without go-between intercession. Whereas companies inevitably target uti-
lizing open blockchains for commerce operations, actualizing private blockchains is
less demanding to control and oversee security and protection. The Blockchain and ERP

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 31
integration makes and ports unused data into the dispersed record and draws existing
information from endeavor frameworks [3]. Blockchain ERP combination empowers
businesses to control information sharing ﬁrmly.
The decentralization property of Blockchain gives companies the potential to join
together a huge supply chain organize. Blockchain in ERP combinedly makes a differ-
ence to make strides in the robotization supply chains. So, each company can utilize
its inside ERP framework whereas counting one rule-enforced blockchain arrange. This
brings up a few unmistakable focal points. For illustration, it diminishes the visibility
issue inside the supply chain. Each member within the whole framework can follow the
item travel from fabricating to last conveyance on the rack.
Not at all like ERP frameworks, Blockchain ERP arrangement is additionally cheaper
as all the members like producers, retailers, and wholesalers can utilize the same arrange-
ment. So, the trade rules and the information right now dwelling with diverse compa-
nies would be accessible in one organization [1]. And each exchange would naturally
be recorded on the dispersed record innovation. Companies would have perceivability
into the complete supply chain. In any case, they can still record and keep the private
information important. They don’t have to be share commerce insights.
Last but not least, having transparency of the data from the ERP system of the e-
commerce companies, makes possible the economic forecast and the accurate estimation
of the company’s statistical forecasts.
5 Conclusions and Future Research
In this paper, I have discussed what blockchain technology has brought to an e-commerce
company, more precisely at the ERP level, including excellent security performance,
helped to reduce frauds, improved performance at the SCM level, and improved eco-
nomic forecasts due to transparent data brought through the smart contract, but it also
brings high energy costs. Considering the DCS theorem [21], energy costs in enter-
prise permissioned BC platforms are signiﬁcantly lower compared to pure decentralized
solutions like Ethereum or Bitcoin which employ Proof of Work as the consensus mech-
anism. The energy cost is blurred by the proﬁt from the implementation and integration
of this solution, as the performance and security costs increase signiﬁcantly, and with
this slope, the customer’s conﬁdence and gain increases.
I have also proposed an architecture model at the ERP level, more precisely for better
supply chain management, by integrating the Apache Kafka architecture with Blockchain
technology, a more secure payment method that avoids online fraud, and a model for an
ideal solution for creating real customers, starting from the proposed intelligent contract
integrated with machine learning for the review section in an electronic store. I analyzed
the beneﬁts and costs of implementing the model within the ERPs of the e-Commerce
stores in Romania, eMAG, Fashion Days, and Altex, which are based on a traditional
ERP, but not much different from the other. Thus, it was found that the change brings the
maximization of the number of customers through the numerous performances brought
to the IT system. It is a cost-effective decision to adopt this solution, compared to the
model that Amazon has adopted, which is mentioned in Sect.2of this paper.
Figure5shows that the use of the blockchain in an ERP-type system may bring
excess costs but compensates for the fact that the performance of the system increases

32 M. G. Stoica
signiﬁcantly, this result having the effect of satisfying the customer with a secure, trust-
worthy product. For the widespread adoption of smart contracts by e-commerce markets,
blockchain companies should ensure that they reduce the operational complexity and
costly transit process of new technology with an existing traditional ERP system.
Fig. 5.Technology cost correlated to the performance of ERP system
As future developments, I would like to pursue the application of this technology at
the level of CRM and how it could improve the entire business logic of the company
because the blockchain provides huge access to data through its transparency at the
system level.
Acknowledgement.This paper was co-ﬁnanced by The Bucharest University of Economic
Studies during the Ph.D. program.
References
1. Faulkner, S.: Beneﬁts of Blockchain in E-commerce, case studies, and resources (2021)
2. 5 Ways Blockchain is transforming eCommerce.https://hackernoon.com/5-ways-blockchain-
is-transforming-ecommerce-b90b4dd61f1c. Accessed 26 Jan 2022
3. Treiblmaier, H., Sillaber, C.: The impact of blockchain on e-commerce: a framework for
salient research topics. Modul University Vienna, Am Kahlenberg (2021)
4. Stallone, V., Wetzels, M., Klaas, M.: Applications of Blockchain Technology in market-
ing—A systematic review of marketing technology companies. Zürich University of Applied
Sciences, Winterthur, Switzerland (2021)
5. Banerjee, A.: Blockchain technology: supply chain insights from ERP, principal consultant,
Chap. 3. Infosys Limited, India (2018)
6. Sundarakani, B., Ajaykumar, A., Gunasekaran, A.: Big data-driven supply chain design and
applications for blockchain: an action research using case study approach. The University of
Wollongong in Dubai, United Arab Emirates (2019)
7. Sunmola, F., Burgess, P., Tan, A.: Building blocks for blockchain adoption in digital
transformation of sustainable supply chains. University of Hertfordshire, UK (2021)

Smart Contracts: A Valuable Technology in ERP e-Commerce Systems 33
8. Consumer satisfaction analysis with PLSPM.https://help.xlstat.com/s/article/consumer-sat
isfaction-analysis-in-excel-with-plspm. Accessed 02 Feb 2022
9. 3 Immediate Blockchain uses for Ecommerce Companies.https://www.practicalecommerce.
com/3-immediate-blockchain-uses-for-ecommerce-companies. Accessed 22 Jan 2022
10. Pew Research Center.https://www.pewresearch.org/fact-tank/2020/03/05/most-americans-
rely-on-their-own-research-to-make-big-decisions-and-that-often-means-online-searches.
Accessed 04 Feb 2022
11. World Economic Forum.https://www.weforum.org/platforms/shaping-the-future-of-techno
logy-governance-blockchain-and-distributed-ledger-technologies. Accessed 12 Jan 2022
12. Amazon SCM-Blockchain.https://aws.amazon.com/blockchain/blockchain-for-supply-
chain-track-and-trace/. Accessed 20 Jan 2022
13. E-commerce in Romania, Statistics and Facts.https://www.statista.com/topics/7418/e-com
merce-in-romania/#dossierKeyﬁgures. Accessed 03 Jan 2022
14. Solana Pay Gateway.https://www.analyticsinsight.net/solana-pay-a-payments-platform-that-
will-overtake-paypal/. Accessed 21 Mar 2022
15. Conversion Rate.https://adoric.com/blog/what-is-a-good-conversion-rate-2020/. Accessed
21 Mar 2022
16. Banerjee, A.: Blockchain Technology: Supply Chain Insights from ERP. Infosys Limited,
India (2018)
17. Kumar, S.A., Prasath, K.: Smart contracts and supply chain management using blockchain.
J. Eng. Res. (2022). ACMM Special Issue, India
18. Karamchandani, A., Samir, K., Rajiv, K.: Perception-based model for analyzing the impact
of enterprise blockchain adoption on SCM in the Indian service industry. Int. J. Inf. Manag.
India52, 102019 (2019)
19. Karamchandani, A., Samir, K., Kumar, S., Srivastava, A.: Analysing perceived role of
blockchain technology in SCM context for the manufacturing industry. Int. J. Prod. Res.
59, 3398–3429 (2021)
20. Al-Farsi, S., Muhammad, M.R., Bakiras, S.: Security of blockchain-based supply chain
management systems: challenges and opportunities. MDPI, Basel, Switzerland (2021)
21. Slepak, G., Petrova, A.: The DCS theorem (2018).https://doi.org/10.48550/arXiv.1801.043
352018

A Nursing Mobile Robot for COVID19 Patients
Using a RPI and Mobile Device
Víctor Lomas-Barrie
1,2,3,4
, Mario Peña-Cabrera
1
, Mauricio Reyes-Castillo
2
,
T. Alcantara-Concepcion
3
, and Gastón Lefranc
4(B)
1
Departamento de Ingeniería en Sistemas Computacionales, IIMAS, Universidad Nacional
Autónoma de México, Mexico City, Mexico
{victor.lomas,mario.pena}@iimas.unam.mx
2
Laboratorio de Robótica, CIDI, Universidad Nacional Autónoma de México, Mexico City,
Mexico
[email protected]
3
Dirección General de Cómputo y Tecnologías de la Información y Comunicación, Universidad
Nacional Autónoma de México, Mexico City, Mexico
[email protected]
4
Pontiﬁcia Universidad Católica de Valparaíso, Valparaíso, Chile
[email protected]
Abstract.The article shows the design and implementation of a caterpillar struc-
ture mobile platform, which includes integrated electronic communication and
control devices, as well as 3D printed parts for custom mechanical and additive
manufacturing. The objective is to help in the care of patients with COVID19
within the restricted areas of the hospital. PWM technique is used for motion
control due to its proven use of robust and reliable performance for service robot
applications. The article shows the design and implementation of a caterpillar
structure mobile platform, which includes integrated electronic communication
and control devices, as well as 3D printed parts for custom mechanical and addi-
tive manufacturing. The objective is to help in the care of patients with COVID19
within the restricted areas of the hospital. PWM technique is used for motion
control due to its proven use of robust and reliable performance for service robot
applications. The implementation is done with a commercial mechanical cater-
pillar platform with the architecture and electronic components manufactured by
3D printing are described in order to show the scope and technique used to con-
trol the operation. The article aims to test the efﬁciency and reliable performance
of a mobile platform using PWM control techniques and Wi-Fi communication
implemented in a RaspBerry Pi hardware electronic device as an auxiliary nursing
robotic system in hospital areas restricted by COVID19. Assisted remote control
is web-based, so any mobile device can perform actions. MQTT WebSocket is
used to send and receive commands and telemetry.
Keywords:Mobile robot·Nursery·COVID19·3D manufacturing·MQTT·
Wi-Fi
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Dzitac et al. (Eds.): ICCCC 2022, AISC 1435, pp. 34–46, 2023.
https://doi.org/10.1007/978-3-031-16684-6
_3

A Nursing Mobile Robot for COVID19 Patients Using a RPI and Mobile Device 35
1 Introduction
The global and digitized society today ﬁnds in mobile robotics a ﬁeld of development
in continuous growth with important expectations in the near future. Trends such as
the delivery of packages with drones by Amazon and DHL, or the granting of licenses
for autonomous cars are proof that every day robotics is and will be more present in
contemporary everyday life. However, there are applications that by their nature need to
be very robust and require the complementary intervention of a human due to the risk
involved, whether to interact with human lives, living beings or in high-risk areas, this
makes the mobile robotics human assistance systems a very powerful tool.
Mobile robots are used for many purposes, but one of particular relevance, due to the
COVID19 Pandemic, is to assist health personnel in hospitals with patients infected with
the SARS-CoV-2 virus. Basic tasks performed by a mobile platform such as carrying
food, medicine and establishing a remote communication with a patient is the subject
of this article. This paper deals with the design and implementation of a commercial
caterpillar platform to provide it with a remote-control system and visual feedback for
its movement. This conﬁguration allows to control such a robot system with greater
precision and smoothness movements, as well as to guarantee that the distance travelled
and the direction of the mobile is correct by means of a remote-control system through
Wi-Fi networks. Which guarantees to operate correctly [1].
The crisis caused by the pandemic due to Covid-19, researchers from developing
countries such as Mexico and Chile, are dedicated to ﬁnding local solutions, which are
shared, as a way to relieve collapsed medical systems. There are projects similar to the
one described in this document, such as the development of a kiosk-platform called
Charlie in 2014 focused on helping to monitor the vital signs of older adults in hospitals.
The mobile robot was developed at the Korean company Yujin Robotics, with the
initial goal of serving in restaurants and supporting teachers in schools. Charlie is 1.2 m
tall, has a 24V LiPo battery and is made up of a camera, a touch screen, speakers,
microphones, contact and laser sensors, and two computers. The robot adapts with more
components or different software depending on the environment it is in; therefore, it is
possible to add a sphygmomanometer, oximeter, glucometer, or more cameras and micro-
phones. It is described that it also has blood pressure measurement and an entertainment
module with music, videos and games [2].
The Smooth project has been developed in Denmark since 2017, with a prototype that
has been designed to support the transport of materials for older adults; is a developing
robot with three goals: to be a garbage or laundry collector, to serve drinks, and to be
a guide, walker, or wheelchair that supports the movements of the adult in action [3,
4]. In [5], a mobile robot is incorporated, assistant to the medical staff, who cares for
older adults. In this case, PR2 has the ability to manipulate objects; with a retail price of
USD 400,000 [6]. The study contains the opinions of 14 nurses, who expressed fear of
being replaced by robotic assistants, and who prefer the robot to perform support tasks
to save time, such as having a third hand. In this same sense, other studies were found,
such as the one by Turja et al. in 2018 [7]. The autonomous service robot for hospitals
(Relay) of (Swisslog) [8], used for the transport of medicines, laboratory samples and
other medical materials; and uses laser navigation, computer vision, and has a touch
screen and wireless communications.

36 V. Lomas-Barrie et al.
On the other hand, applications have been reported in telemedicine [9] and with
the aim of promoting sustained research in robotics to address infectious disease risks
[10]. Finally, a study carried out in 2017 [11] addresses the problem by determining
the characteristics to consider when designing a nursing assistant robot, similarly to the
situation presented in 2020 with COVID-19, the study was carried out after detecting the
needs of patients and nurses in the face of MERS infection in 2015 in South Korea. From
this situation, the need to develop a support system for nursing staff became evident.
Several aspects should be considered based on these works: the mobile robot should
have a pleasant and simple interaction, with dimensions that must be in accordance with
the place where it moves, and be able to navigate avoiding obstacles, and be in harmony
with the environment, for example, it should be considered that the lights, the noise it
produces, the temperature and the alarms do not disturb the environment in which it is
located.
The objective of this paper is to help in the care of patients with COVID19 within
the restricted areas of the hospital.
2 The Mobile Robot’s Description
Mobile robots have characteristics that make them suitable for a certain task, with a
physical structure with traction and steering systems that are in accordance with the
requirements of speed, maneuvers and terrain characteristics. Robot sensors and a control
system, with obstacle avoidance, are added to achieve the precision and speed necessary
to perform a reliable navigation to go from a point to its destination. This implies that the
mechanical, electronic and computer systems must be more complex. There are three
basic systems from which different conﬁgurations can be obtained: drive and steering
on independent axle, drive one direction on the same axle and steering and drive on all
axles [12]. There are three basic systems from which different conﬁgurations can be
obtained: drive and steering on independent axles [13], drive and steering on the same
axle [14] and drive and steering on all axes [15].
In the mobile robot described in this document, a pair of tracks with independent
motors are used, which resembles the traction and steering system in the same axis, also
called differential traction. It must have the necessary information to allow the mobile
robot to carry out three fundamental tasks: estimate its position and orientation, keep
the map of the environment updated and detect possible obstacles.
For the estimation of the position and the updating of the environment map, charac-
teristics such as precision, spatial resolution and range are required, in the detection of
obstacles, the time between observations must be less, to have the information already
processed as quickly as possible. In the case of this robot-nurse project, the above char-
acteristics are not essential since the navigation operation is controlled remotely by a
human operator who has visual feedback through a camera located on the platform itself,
this is due to the type of application and the rapid design and construction required to
have the impact in the time required by the COVID19 Pandemic, in this way the con-
trol of the robot is easier and requires less technical knowledge to be operated within a
hospital environment in an area conﬁned for patients with COVID19.
The mobile robot Ayúdame 1.0 is a platform for the support and location of Food,
Medication and Equipment (Fig.1), developed jointly by the Research Institute in

A Nursing Mobile Robot for COVID19 Patients Using a RPI and Mobile Device 37
Applied Mathematics and Systems (IIMAS), PUCV and the Industrial Design Cen-
ter, both from the National University of Mexico (UNAM). The objective is to serve as
an assistant in nursing processes in conﬁned areas in hospitals with COVID19 patients
in a tele-operated assisted way up to 50 [m]. The design must be practical, robust, stable
and capable of carrying a hospital-type tray with food, medicine and other necessary
materials. In addition, the platform will be able to bring the means to establish a video
call with the patient. Navigation is visual by means of a camera that transmits to a mobile
device, in addition to other telemetry data, including an electronic compass.
The Ayúdame 1.0 system has the following parts: An electromechanical system, an
electronic system, a control system, remote command, support and serving tray.
2.1 The Electromechanical System
The mobile robot Ayúdame 1.0 uses the caterpillar-type commercial platform (T’Rex
Tank) with differential traction based on six support wheels and one traction on each
side and with a total load capacity of up to 8 [kg].
Fig. 1.The mobile robot Ayúdame 1.0
The crawler drive allows to control the movements with greater precision and smooth-
ness, as well as guarantee that the distance traveled and the direction of the mobile is
correct by means of a remote-control system.
The chassis is made of anodized aluminum, and the sprocket wheels, the bearing
wheels, the track and the suspension are made of anodized zinc. The width is 265 [mm]
and the length is 355 [mm]. The mobile structure has independent suspension to improve
traction and absorb shocks, it has a height of 70 [mm] necessary to make the chassis very
operational and manageable on rough terrain and not rub against the ﬂoor. The chassis
has a cavity to transport batteries and components, while its lid serves as a base for
transport or to integrate other mechanical components, which in the case of Ayúdame
1.0 is a structure for transport hospital-type trays to carry food, medicine or some material
necessary for the patient.
The tracks are driven by 2 independent 12 [V] direct current motors with a nominal
current consumption without load of 1.3 [A], with load of 4 [A] and stall of 11 [A].
Both motors are controlled by a STMicroelectronics VNH5019 dual H-bridge and the
electrical power is supplied by a lithium-ion polymer (LiPo) 3S35C battery of 5,200
mAH.

38 V. Lomas-Barrie et al.
2.2 Electronic System
Figure2shows a connection diagram of the different components that make up the control
system of the mobile robot AYUDAME 1.0. The main component of the electronic
system is the small format computer, the Raspberry Pi 4 (RPI) with 4 GB of RAM.
The communication processes, control of the caterpillar traction motors, reading of the
current consumed by the motors, reading and processing of values coming from an
inertial measurement unit, are carried out by the RPI, as well as the detection of objects
and surfaces to through two ultrasounds. Sensors and panning movements of the camera.
The operating system used is Raspbian 10 (Buster) with Linux Kernel 4.19 in its Lite
version.
Fig. 2.Diagram of electronic components
Port and pin monitoring is described in Table1. General Purpose Port (GPIO) mon-
itoring is done through the Python library (Gpiozero). Three PWM channels are used,
one of them implemented by software is to control the panning of the servo motor (SM)
that moves the camera, the other two DMA channels are used to ensure the maximum
frequency of 20 [kHz] to regulate the speed of the camera. VNH5019 circuit, which
controls the traction motors (M1 and M2). 3 pins are needed to control the direction and
enable of each motor (cM1 and cM2).
The current consumed by the motors (cM1 and cM2), the VNH5019 dual IC provides
two analog channels, one for each H-bridge, which are read by the ﬁrst and second
channels of the MCP3008 analog-to-digital converter. Then, they are read sent by the
RPI through the I2C protocol. The reading of the IMU AltIMU-10 electronic compass
is captured on this same channel. Two GPIO pins are used to drive two SRF05 sensors
(left, I-SRF05 and right, D-SRF05) and two to listen for echo return to detect an object
in front of the platform. The platform stops when an object is detected within a distance
of 15 [cm] and a signal is constantly sent to the remote control to avoid collisions. The
battery is connected to a regulated power supply (D24V25F5) that powers the Raspbery
Pi, the AltIMU-10 and the MCP3008 with 5 [V]. An emergency stops button (SWPARO)
and a general switch (SWGRAL) cut the power supply when necessary.

A Nursing Mobile Robot for COVID19 Patients Using a RPI and Mobile Device 39
Table 1.Port pin assignment for RPI
Port(pin) Type Device
GPIO[2, 3] I2C MCP3008
GPIO[2, 3] I2C AltIMU-10
GPIO[23, 24] I/O D-SRF05
GPIO[22, 27] I/O I-SRF05
GPIO[17] PWM SM
GPIO[12, 13] PWMDMA VNH5019
GPIO[5, 6, 14, 15, 16, 18] I/O VNH5019
CSI BusCSI Camera
2.3 Control Systems
The architecture of the control system of the mobile platform Ayúdame 1.0 is imple-
mented in the RPI software and has a remote control (Fig.3). The Control Server is
developed in Python 3 with the purpose of managing address commands and telemetry
data propagation to and from the control motors M1 and M2, and SM, as well as current
draw feedback, reading of SRF05 sensors and the AltIMU. -10. It sends and receives
data through a broker-provided MQTT service installed and running on the RPI using
WebSockets over port 9001, as the remote uses JavaScript. (Figure3) shows the Wi-Fi
network and the set of ﬁles loaded by the remote-control browser.
Fig. 3.Control software architecture
Risk mitigation actions prevent the Wi-Fi signal from being weak, the platform
automatically stops when a deﬁned threshold is reached, and a visual alert is displayed
on the remote.

Random documents with unrelated
content Scribd suggests to you:

Krause that it was very rarely employed. See Die Byzantiner des Mittelalters, by J.
H. Krause; Halle, 1869.
294 Pusculus, iv. 340.
295 Phrantzes.
296 Gyllius mentions this foreshore as existing in his time, gives its width, and
vividly describes how it was utilised and increased by the inhabitants of Galata
(book iv. ch. 10). In digging for the foundations of the British post office in Galata
in 1895, on a site that is now upwards of a hundred yards from the water, remains
of an old wooden jetty were discovered. Indeed, I think it highly probable that in
1453 the whole of what is now the main street of Galata from the bridge to
Tophana was under water.
297 Pusculus, 247.
298 Crit. xli.
299 Barbaro, p. 24, and Phrantzes.
300 According to Ducas, Mahomet himself inflicted the blows: an absurd
statement.
301 Ducas, 121; Leonard, Phrantzes, and Nicolo Barbaro.
302 Hunyadi, according to Phrantzes (p. 327), asked that Silivria or Mesembria, on
the bay of Bourgas, should be given to him as the price of his aid, and Phrantzes
declares that the emperor ceded the latter place, he himself having written the
Golden Bull making the cession. He adds also that the king of Catalonia stipulated
for Lemnos as the price of his aid. But no aid came from either.
303 Barbaro, under April 21; Phrantzes, 246. The tower is called by Leonard
Bactatanea. He afterwards writes of the breach near it as being in the Murus
Bacchatureus. See, as to its situation, Professor van Millingen’s Byzantine
Constantinople, pp. 86, 87.
304 As the only church in the neighbourhood of the place defended by Justiniani
was that of St Kyriakè near the Pempton, the information is valuable as helping to
fix the locality where the great gun was stationed. The Moscovite, ch. vii.
305 The Moscovite, ch. vii., in Dethier’s Siege; Barbaro, p. 27; Crit.
306 Zarabotane.
307 Barbaro, p. 27. The account of the fight given by Pusculus is very full and
spirited. See note in Appendix as to the question where the naval fight took place.
308 In 1203 the Crusaders and Venetians had forced the boom tower on the
Galata side and loosed the chain; but it was then outside the city walls. In the
time of Cantacuzenus, Galata had been enlarged so that the end of the chain was

quite safe unless Galata were taken. The walls terminated, as may still be seen by
the remaining towers, near Tophana.
309 Leonard, and Sauli’s Colonia dei Genovesi in Galata, p. 158. Other similar
instances are cited by contemporaries, but it is not necessary to suppose that
Mahomet had ever heard either of the fable of Caesar’s attack upon Antony and
Cleopatra or of a like feat performed by Xerxes. The Avars had made a crossing
similar to that contemplated by Mahomet. The transport of the imperial fleet into
Lake Ascanius in order to take possession of Nicaea in 1097 might possibly have
been known to him.
310 Λοιπὸν ὁ ἀμερᾶς τὰς τριήρεις φέρας ἐν μίᾳ νυκτί, ἐν τῷ λιμένι τῷ πρωῒ
ηὑρέθησαν: Phrantzes, 251.
311 Dethier places them on a small plateau now occupied by the English Memorial
Church. [Note on Pusculus, book iv. line 482. Professor van Millingen (p. 231), in
discussing the question of the position of St. Theodore, suggests that the sultan’s
battery stood nearer the Bosporus than the present Italian Hospital. This
suggestion is not necessarily at variance with the position indicated by Dethier.]
312 Philelphus, book ii. line 976: ‘Genuae tunc clara juventus obstupuit.’ Ducas,
however, states that the Genoese claimed to have known of the proposed
transport and to have allowed it out of friendship to Mahomet.
313 ‘Et hic quidem in superiori parte per montem navigia transportavit ... in litore
stabant milites parati propulsare hostes bombardis, si accederent prohibituri
deducere naves.’ Chalcondylas, book viii.
314 Crit. says 68; Barbaro, 72; Tetaldi, between 70 and 80; Chalcondylas, 70; and
Ducas, 80; Heirullah says there were only 20; the Janissary Michael, 30; the Anon.
Expugnatio, edited by Thyselius, sect. 12, says not less than 80.
315 ‘Lacertus’ is the word Leonard ingeniously uses for the Greek πῆχυς.
316 Crit. book iv. ch. 42. It is difficult to determine the size of the boat selected for
this overland transit. Barbaro says, ‘le qual fusti si iera de banchi quindexe fina
banchi vinti et anchi vintido’ (page 28). This would agree fairly well with the
statement of Chalcondylas, that some had thirty and some fifty oars. Mr. Cecil Torr
calculates that a thirty-oared ship would be about seventy feet long, a statement
which appears probable (Ancient Ships, p. 21). The mediaeval galleys and other
large vessels propelled by oars differed essentially from those of the sixteenth
century, which were worked with long oars. See note on p. 234. I am myself not
entirely satisfied that among the boats were not biremes and possibly triremes in
the sense of boats which had two or three tiers of oars, one above the other.
Fashions change slowly in Turkey, and I have seen a bireme with two such tiers of
oars on the Bosporus. No writer mentions the length of the vessels which were
carried across Pera Hill. A large modern fishing caique in the Marmora, probably

not differing much in shape from the fustae then transported, and containing
twelve oars, measures about fifty feet long. When the boats are longer, two men
take one oar, but this is very unusual. Leonard speaks of the seventy vessels as
biremes. Barbaro calls them fustae. The former was probably the best Latin word
to signify the new form of vessel. Many of the ships were large, though it may be
taken as certain that none were of the length of the two galleys recently raised in
lake Nemi, near Rome, which belonged to Caligula, each of which is 225 feet long
and 60 feet beam.
317 See note in Appendix on transport of Mahomet’s ships.
318 Ducas, xxviii.
319 Phrantzes, p. 327.
320 Crit. lxxii.
321 Barbaro says that the meeting was in St. Mary’s; but Pusculus (iv. 578) says,
in St. Peter Claviger, which Dethier places near St. Sophia.
322 Phrantzes, 256.
323 Barbaro, under April 24 and 25.
324 Pusculus, lines 585 et seq.
325 Pusculus, iv. 610.
326 Barbaro, 31.
327 The account of this attempt to destroy the Turkish ships in the harbour is best
given by Barbaro, but Phrantzes and Pusculus are in substantial agreement with
him.
328 Phrantzes (p. 248) says 260 Turkish prisoners were executed.
329 The Moscovite, ch. vii.
330 Crit. xliv.
331 Dr. Mordtmann places the bridge between Cumberhana and Defterdar Scala.
332 Ducas gives the above dimensions. Assuming the width from centre of each
barrel, including a space between them, to be four feet, this would give the length
of the bridge as 2,000 feet, which is about the width of the Horn at the place
mentioned. Phrantzes gives its length at a hundred fathoms and the breadth fifty
fathoms. These dimensions are clearly wrong if applied to the bridge, since the
length falls far short of the width of the gulf. Leonard says it was thirty stadia long.
Here, as elsewhere, I suspect that he uses stadium for some measure about one
ninth of a furlong in length. If this conjecture is right, his estimate of the length of
the bridge is about 2,000 feet.

333 Phrantzes, 252.
334 Barbaro, 36; Phrantzes, 250.
335 The Moscovite, xv. While there are useful hints in this anonymous author, he
is generally untrustworthy. This fight, for example, is represented as being outside
the walls. It is incredible that the Greeks should have made a sortie at this period
of the siege. As an illustration of the untrustworthy character of the writer, it may
be noted that the number of Turks killed during the siege totals up to 130,000!
336 Leonard, the Vallum and the Antemurale.
337 Phrantzes, p. 244.
338 ‘Bastion’ is the word used for a wooden tower or castle by Barbaro and by the
translator of the Moscovite. Chalcondylas calls it helepolis, distinguishing it from
the cannon which he names teleboles. Ducas speaks of cannon usually by the
word χωνείαν, sometimes as τὰς πετροβολιμαίους χώνας or σκευαὶ πετροβόλοι or
simply as τὸ σκεῦος; Phrantzes employs the word helepolis for a wooden turret
(pp. 237, 244). The latter word is used by Critobulus for a cannon. It was an
epithet applied to Helen, ‘the Taker of Cities.’ In the Bonn edition of Phrantzes it is
also employed, both in the text and the Latin translation, for cannon; but a
reference to the readings of the Paris MS. suggests that it is an error. Phrantzes’s
words for cannons are teleboles and petroboles.
339 The ‘Chastel de bois’ was ‘si haut, si grand et si fort qu’il maistrisoit le mur et
dominait par-dessus’ (Tetaldi, p. 25).
340 Barbaro states that it occupied a place called the ‘Cresca,’ possibly a copyist’s
error for Cressus (= Chariseus), the name which I believe he gave indifferently
with San Romano to the Pempton. Elsewhere he uses Cresca for the Golden Gate
(e.g. p. 18). Possibly, however, he is referring to another turret, which was at the
Golden Gate. Barbaro’s knowledge of places and names is not accurate. If
Barbaro’s ‘bastion’ is the ‘helepole’ of which Phrantzes speaks (p. 245), then the
three writers agree that the principal turret was at the Romanus Gate.
341 The Moscovite, 1087; Phrantzes, 247.
342 Leonard, p. 93: ‘Mauritius Cataneus ... inter portam Pighi, id est fontis, usque
ad Auream contra ligneum castrum, pellibus boum contectum, oppositum accurate
decertat.’ Cardinal Isidore, in the Lamentatio, says, p. 676: ‘Admoventur urbi
ligneae turres.’
343 Barbaro, under dates of May 21, 22, 23, 24, and 25.
344 As to the question whether there was water in the foss, see Professor Van
Millingen’s Byz. Constantinople, pp. 57–8.
345 Crit. xxxi. Ἀλλὰ τοῦτο μὲν ὕστερον περιττὸν ἔδοξε, καὶ ματαία δαπάνη, τῶν
μηχανῶν τὸ πᾶν κατεργασαμένων.

346 The return, as mentioned, was on May 23, but is given by Barbaro under the
3rd. This is one of the passages which show that his diary was revised and added
to after the siege.
347 Crit. xlvi.; Pusculus, iv. 889, says:
Candida completo cum Phoebe surgeret orbe
Moesta prodit, fati miseri cladisque propinquae
Nuntia; nam tristis faciem velamine nubis
Tecta atrae, mediaque latens plus parte sereno
Incedit coelo.
Barbaro seems to describe an eclipse of the moon on May 22. The elder Dr.
Mordtmann states that there was no full moon and consequently no eclipse on the
22nd, but that there was on the 24th. Dethier’s note on The Moscovite, p. 1100.
Phrantzes, p. 264, speaks of a light flashing from the sky settling over the city, and
remaining during the whole night. See note, post p. 316.
348 Constantine was a widower, his wife, Catherine, having died in 1442, a year
after her marriage. Phrantzes, 195–8.
349 The same remark applies to The Moscovite generally. There are so many
manifest fringes to what ought to have been the correct narrative of an eye-
witness that it is impossible to distinguish truth from falsehood.
350 Barbaro, under May 20.
351 Leonard, Opere, p. 94.
352 Leonard, p. 92.
353 Ibid. p. 95.
354 Barbaro, under May 28.
355 Ep. Ang. Johannis Zacchariae Potestatis Perae, Sec. 2, edition revised by
Edward Hopf and Dethier.
356 Leonard, p. 94, and also Italian version given by Dethier, p. 644.
357 Tetaldi, pp. 32–35.
358 Crit. xlviii.
359 See also the Moscovite, xx.
360 Crit. lx.
361 Barbaro, Pusculus, and Leonard agree with Critobulus in their description of
the stockade.
362 Phrantzes, 263.

363 Ibid. 326. M. Mijatovich, in his pleasant and valuable Constantine, last
Emperor of the Greeks, states that Mahomet received an ambassador from
Ladislaus on May 26 (p. 198); but I do not know on what authority.
364 Phrantzes, 325.
365 M. Mitjatovich’s suggestion that the negotiations had probably emanated from
the wily cardinal who had been the evil spirit of Ladislaus, or possibly from the
crafty, but unpractical, mind of George Brancovich, appears plausible.
366 Phrantzes, 326; Ducas, xxxviii.
367 Ducas, xxxviii.
368 Tetaldi says: ‘Se l’armée de Venise que menoit et conduisoit Messire Jean le
Rendoul [Loredano] fut arrivé à Constantinople ung seul jour avant que cette cité
fust prinse, certes il n’y avoit aucun doute qu’ils eussent fort secouru et fussent
venus bien à point’ (p. 30).
369 ‘Per el campo del Turco in questo zorno se fexe asai feste, de soni, e de altra
condition de alegreze, e questo perche i sentiva che tosto i volea dare la bataia
zeneral’ (p. 48, under May 24).
370 Phrantzes, 263.
371 Leonard, p. 95; Phrantzes, 263; Crit. xlvi.
372 Crit. xlvii.
373 The accounts of this light (or darkness), which alarmed both sides, are
somewhat conflicting. Perhaps here also Critobulus is the safest guide. In chapter
xlvi. he mentions the religious procession already described, where the statue of
the Virgin falls, and says it was ‘three or four days before the attack.’ Immediately
after came torrential rains with vivid flashes of lightning. Then, ‘the next day,’
there was a thick fog lasting till evening. Barbaro speaks of a darkness, due,
judging from his description, to an eclipse of the moon, lasting from the first to
the sixth hour after sunset, as being on the 22nd. This alarmed the Greeks, he
says, because of an ancient prophecy which declared that Constantinople should
not be lost until the moon should give a sign in the heavens. Phrantzes (page 264)
says: φῶς ἀστράπτον καταβαῖνον ἐξ οὐρανῶν καὶ δι’ ὅλης τῆς νυκτὸς ἄνωθεν τῆς
πόλεως ἑστὸς διέσκεπεν αὐτήν. Possibly both Phrantzes and Barbaro have the
same atmospheric night effects in view: that is, that there were frequent flashes of
lightning during the night so long as the eclipse lasted. The statement of Pusculus,
who was in the city at the time, has already been quoted. See p. 297, ante. The
account of Critobulus appears clear, but it does not eliminate the miraculous, for
he declares that many persons, both Romans and foreigners, declared that they
had seen the Divinity hiding Himself in the clouds.

374 Ducas also mentions the attempt recorded by Chalcondylas, but without
mentioning the name of Ismail. Ducas thus mentions two negotiations for peace,
the first (if it ever existed) being towards the end of April and the second nearly a
month after.
375 The Turkish historian Sad-ud-din, (p. 20) represents the emperor as offering
to surrender everything except Constantinople; to which Mahomet’s reply was,
‘Either the city, the sword, or El-Islam.’
376 Leonard.
377 Leonard, Phrantzes, and Tetaldi all speak of him as friendly to the Christians.
He was, however, disliked by Mahomet, because he had persuaded Murad to send
his son to Magnesia. Tetaldi says that the Christians in the Turkish army shot
letters into the city to let the besieged know all that went on in the council.
378 According to Leonard, the sultan ordered Zagan to fix a day for a general
assault.
379 Phrantzes, 623–8, and also Leonard.
380 The narrative of Phrantzes relating the decision of the meeting of the Turkish
council concludes by stating that this was on the 27th—that is, Sunday (p. 269). It
may have been, but it is difficult to believe that the council meeting, the sending
of Zagan to learn the opinion of the soldiers, his return and the decision, together
with the subsequent proclamation, were all crowded into one day. Barbaro gives
the proclamation as being made on Monday the 28th. Leonard says that, as a
result of the meeting, a proclamation was issued for the attack to be on Tuesday
and for the three preceding days to be devoted to prayer and one of them to
fasting. If he is correct, the council could not have been on the 27th. Tetaldi states
that the council lasted during four days. The statement appears possible, and
perhaps gives the explanation of the apparent discrepancies in the narratives.
381 Leonard, 96, Phrant. 269; Barbaro adds that the Turks believed that on the
morrow they would have so many Christians in hand that two slaves could be
bought for a ducat: such riches that everything would be of gold, and they could
have enough hair from the heads of Christian priests to make ropes with which to
tie up their dogs.
382 The Moscovite, xxii. This first wound is only mentioned by the Moscovite.
383 Phrantzes, 269.
384 Barbaro, p. 50.
385 Barbaro. Ducas says, from St. Eugenius to Hodegetria and as far as Vlanga (p.
282–3), which is substantially the same position as that given by Critobulus.
386 Zorso Dolfin, p. 78.

387 Sad-ud-din, p. 16. Translation by E. J. W. Gibb.
388 τούφακας; in modern Greek the name for sporting guns is τουφέκια. The
Turks call them Toufeng. Ducas uses the word μολυβδοβόλοι.
389 Crit. xlvii. to lii.
390 According to Critobulus, the meeting of the Council was on the 27th.
391 Phrantzes, 269–70. Was the speech as recorded by Critobulus ever delivered?
The answer I am disposed to give is that a speech was delivered which was
substantially that reported by Phrantzes and Critobulus. The fashion followed by
the Byzantine writers, and their desire to imitate classical models, by putting all
speeches in the first person, made it necessary to invent a speech if the substance
of what was said were known. Critobulus, writing some years after the capture
and having had many opportunities of meeting with the Turkish leaders, was in a
position to learn what was said and done by them, and hence his report, wherever
it can be tested, almost invariably proves trustworthy.
392 Barbaro, May 28.
393 Crit. liv.
394 Phrantzes, 271–8; Leonard, 97.
395 Phrantzes, 279; The Moscovite, p. 1113. The ceremony is also mentioned in
the Georgian Chronicle.
396 Libro d’Andrea Cambini Florentino della Origine de Turchi et imperio delli
Ottomanni. Edition of 1529, p. 25.
397 Phrantzes, p. 280. The closing of the gates behind the soldiers is mentioned
also by other writers.
398 The Caligaria Gate was the present Egri Capou. For a description of Caligaria
and the neighbouring palace of Blachern see Professor van Millingen’s Byzantine
Constantinople, p. 128. Caligaria was the name of a district which was in the
corner made by the wall running at right angles to the foss, where it terminates on
the north just beyond Tekfour Serai, and that which leads down the steep slope to
the Golden Horn.
399 Phrantzes, p. 280.
400 The question when the general attack began is very much one of
appreciation. According to Ducas, Mahomet commenced on the Sunday evening to
make a general attack and during the night the besieged were not permitted to
sleep but were harassed all night and, though in a less active manner, until
between four and five of the afternoon of Monday. Phrantzes declares the capture
to have been made on the third day of the attack and would thus make it begin on
Sunday, but his narrative shows that the general attack began after midnight of

the 28–9th. Barbaro’s statement substantially agrees with that of Phrantzes and is
that during the whole of the 27th the cannons were discharging their stone balls:
tuto el zorno non feze mai altro che bombardar in le puovere mure; but on p. 51
he says that Mahomet came before the walls to begin the general attack at three
hours before day on the 29th. Critobulus makes the general attack begin on the
afternoon of the 28th, when the sultan raised his great standard (Crit. lii. and lv.).
Karl Müller, in his excellent notes to Critobulus, justly remarks that as Barbaro and
Phrantzes were in the city their evidence ought to be preferred to that of
Critobulus. They both represent the final assault as beginning very early in the
morning of the 29th. The statements are reconcilable by supposing that the
dispositions for a general attack began on the Sunday, but that the actual general
assault did not take place until the Tuesday morning. Sad-ud-din says, on the
authority of two Turkish contemporaries, that ‘the great victory was on Tuesday,
the fifty-first day from the commencement of the war’ (p. 34).
401 Cambini, 24.
402 P. 160.
403 Ch. lv.
404 P. 52.
405 Leonard, p. 86: ‘Testis sum quod Graeci, quod Latini, quod Germani, Panones,
Boetes, ex omnium christianorum regionibus Teucris commixti opera eorum
fidemque didicerunt.’
406 Riccherio, 958: ‘Percioche Maometh pensava, ricreando gli stracchi col rimetter
nuove genti nella zuffa, verrebbe a non dar punto di spatio per riposarsi a Greci, di
maniera che, non potendo sostener tanta fatica per lo continuo combattimento, si
sarebbono agevolmente potuti vincere.’
407 Crit. liv.
408 Michael Constantinovich, a Servian who was with a contingent of his
countrymen in the Turkish army, says, ‘As far as our help went, the Turks would
never have taken the city’ (quoted by Mijatovich, p. 234).
409 τούφακας, Crit. li.
410 Chalc. p. 160.
411 Barbaro (54) says, Greeks and Venetians, omitting all mention of the
Genoese.
412 Crit. lvi.
413 Leonard: ‘in loco arduo Myriandri.’
414 Pusculus, iv. 173, and Zorzo Dolfin, 55.

415 Crit. lvii.
416 Leonard, p. 98: ‘Tenebrosa nox in lucem trahitur, nostris vincentibus. Et dum
astra cedunt, dum Phoebi praecedit Lucifer ortum, Illalla, Illalla in martem
conclamans, conglobatus in gyrum consurgit exercitus.’
417 Crit. lvii.
418 Παραπόρτιον ἓν πρὸ πολλῶν χρόνων ἀσφαλῶς πεφραγμένον, ὑπόγαιον, πρὸς
τὸ κάτωθεν μέρος τοῦ παλατίου.
419 Its complete name was Porta Xylokerkou, because it led to a wooden circus
outside the city. See the subject fully discussed by Professor van Millingen,
Byzantine Constantinople, pp. 89–94.
420 I am not satisfied that the Kerkoporta was the one indicated by Professor van
Millingen. On the map published by the Greek Syllogos, as well as in Canon
Curtis’s Broken Bits of Byzantium, a small postern is shown in the wall immediately
south of the tower adjoining Tekfour Serai, and my own recollection is that I saw
this walled-up postern with Dr. Paspates in 1875. The wall itself was pulled down
on the outbreak of the last Turko-Russian war and replaced by a slighter one.
Whichever view be correct, the statement in the text is not affected.
Professor van Millingen contends that the Kerkoporta strictly so called was the
small gate in the corner between Tekfour Serai and the adjoining tower on the
south. But he maintains also that the postern to which Ducas refers was in the
transverse wall, giving access from the city to the Inner Enclosure. He remarks
that if the Turks entered by the Kerkoporta they could have mounted the great
Inner Wall from the city. As to the latter objection, it must be remembered that
the fighters were within the Enclosure defending the Outer Wall, and if the Turks
entered through the postern in the transverse wall they would take the fighters in
the rear. It would have been a better position for attack than on the Inner Wall.
421 Phrantzes, p. 285.
422 Crit. lvi.
423 Sad-ud-din gives an interesting variant of the story of Ducas. He states that
while ‘the blind-hearted emperor’ was busy resisting the besiegers of the city at
his palace to the north of the Adrianople Gate,’ ‘suddenly he became aware that
the upraisers of the most glorious standard of “The Word of God” had found a
path to within the walls’ (Sad-ud-din, p. 30). The statement that the emperor was
present at Tekfour Serai agrees with that of Ducas; but the latter’s account of the
events immediately following the entry by the Kerkoporta varies so much from that
given by others that I suspect some sentences have dropped out of his narrative.
424 Crit. lviii.
425 Ibid.

426 Leonard, p. 37.
427 It is difficult to identify the gate described as having been opened on to the
stockade. Critobulus gives no further indication of its position than that here
mentioned (ch. lx.). Paspates thinks it was a temporary postern, walled up after
the siege when the Inner Wall was repaired to prevent smuggling, but would place
it not far from Top Capou, a position which cannot be accepted if the stockade
were, as I have placed it, near the Military Gate of St. Romanus. The Podestà of
Pera, however, says that Justiniani went ‘per ipsam portam per quam Teucri
intraverunt’ (p. 648), which would indicate St. Romanus. Andrea Cambini, the
Florentine already quoted, in his Libro della Origine de Turchi, published by the
sons of the writer, says that Justiniani, who had behaved so well that the salvation
of the city was largely attributed to him, was seriously wounded, and, seeing that
the blood flowed ‘in great quantity’ and being unwilling that they should fetch a
doctor, withdrew secretly from the fight ... all the gates which led from the
Antimuro [i.e. the Outer Wall] being closed, because thus the fighters had to
conquer or die (p. 25).
428 His monument still exists in the church of S. Domenico at Chios with an
epitaph which contains the phrase ‘lethale vulnere ictus interiit.’ Phrantzes says
that Justiniani was wounded in the right foot by an arrow; Leonard, by an arrow in
the armpit; Chalcondylas, in the hand, by a ball; Critobulus, by a ball in the chest
or throat which pierced through his breastplate. The latter statement would be
consistent with Tetaldi’s which speaks of the wound inflicted by a culverin.
Riccherio says Justiniani was wounded by one of his own men. Barbaro (who, it
must always be remembered where he is speaking of the Genoese, was a Venetian
and incapable of doing justice to a citizen of the rival republic) does not mention
any wound, but states roundly that Justiniani decided to abandon his post and
hasten to his ship, which was stationed at the boom.
429 Barbaro, p. 55.
430 Philip the Armenian, who was probably present in the city, states that
Justiniani and his men deserted their stations and that thus the city was lost (pp.
675–6). Riccherio, while speaking of the wound as severe, declares that Justiniani
promised to return, and attributes the departure of many of his followers to the
fact that the postern gate, which he had required to be opened for his departure,
suggested the idea of flight to his men. In other words it created a panic (p. 960).
The contemporaries who excuse Justiniani are Cardinal Isidore (Lamentatio, p.
677: ‘Ne caeteros deterreret, remedium quaerens clam sese pugnae subduxit’) and
Leonard, who both state that he went away secretly so as not to discourage his
followers. Tetaldi further declares that he left his command to two Genoese.
Leonard and the Podestà wrote while the impression of the fall and the sack of the
city were too recent to enable them to give a cool judgment on Justiniani’s
conduct: the latter dating his letter June 23, and the archbishop August 16.

431 Crit. lx.; also Leonard, 99.
432 Cambini, p. 25.
433 Phrantzes, 285.
434 Crit. lx.
435 Phrantzes, p. 285.
436 ‘La prima sbara di barbacan,’ p. 54.
437 Phrantzes, p. 285.
438 Montaldo, xxiii.: ‘insigniis positis.’
439 Montaldo (ch. xxiii.) incidentally confirms the version of Ducas. He states that
the emperor determined on death only after he had learned that the enemy had
entered the city and had occupied the palace and other places.
440 Leonard, p. 99. In Dethier’s edition a note states that one of the MSS. reads
eighty Latins ‘sine Graecis,’ p. 608.
441 Leonard, 99, says that they formed a cuneus or phalanx.
442 Crit. lxi.; Chalc. p. 164. Ahmed Muktar Pasha’s Conquest of Constantinople.
443 Crit. lxi.; Tetaldi, p. 23, speaks of ‘deux banniers.’
444 Crit. lxi.; Tetaldi, p. 29, ‘à l’aube du jour;’ Barbaro (p. 55) at sunrise.
Phrantzes says that possession of the city was obtained at half past two, which by
the then and present prevalent mode in the East of reckoning time would
correspond to about ten. Possession of the city would probably be about three or
four hours after the entry through the landward walls. Leonard says: ‘Necdum
Phoebus orbis perlustrat hemisphaerium et tota urbs a paganis in praedam
occupatur.’

445 P. 647; ‘on the 29th of last month,’ ‘Qua die expectabamus cum desiderio quia
videbatur nobis habere certam victoriam.’
446 Crit. ch. lxx. Pusculus gives a somewhat different account (iv. 1025):
Auxilium Deus ipse negavit;
In Tenedi portu nam tempestatibus actae
Stabant bis denae naves, quas Gnosia tellus,
Quae Venetum imperium Rhadamanti legibus audit
Omissis, plenas frumento et frugibus, inde
Bis quinas Veneti mittebant Marte triremes
Instructas, urbi auxilio Danaisque; sed omnes
Mensem unum adverso tenuerunt sidere portum;
Nec prius inde datum est se de statione movere
Quam Teucri capiant urbem regemque trucident.
447 Phrantzes, p. 327.
448 Pusc. iv. 1025.
449 Crit. lxxii.
450 Crit. lx.
451 Leonard, p. 99; Polish Janissary, 332; Montaldo notes one report, that he was
trampled down in the throng, and another, that his head was cut off. Philelphus
(book ii. v. 990) says, ‘Enseque perstricto nunc hos, nunc enecat illos, Donec vita
suo dispersa est alma cruore.’
452 See also ch. xxvii. of Montaldo, who adds that the head was sent to the pasha
of Babylon accompanied by forty youths and forty virgins, a procession intended to
make known the sultan’s great victory.
453 The Turks show a place in the bema of St. Sophia which they pretend to be
the tomb of Constantine.
454 Sad-ud-din also makes a Turkish soldier strike off the emperor’s head (p. 31).
455 Phrantzes, p. 291.
456 Until about ten years ago a tomb was shown by local guides to travellers at
Vefa Meidan as the burial-place of Constantine. It bore no inscription. M.
Mijatovich is mistaken in stating (in Constantine, last Emperor of the Greeks, p.
229), on the authority of the elder Dr. Mordtman, that the Turkish government
provides oil for the lamp over his grave. Alongside the alleged grave of
Constantine is that of some one else, probably a dervish, and a lamp was burnt
there some years ago. Similar lamps are burnt nightly in many other places in
Constantinople. It is now entirely neglected. Dr. Paspates suggests, and probably

with truth, that the whole story grew out of the desire for custom by the owner of
a neighbouring coffee-house.
457 ὡς καλὸν ἐντάφιον ἡ βασιλεία ἐστί. The conclusion of Theodora’s speech as
recorded by Procopius.
458 My authority for this statement is on p. 228 of a remarkable book in Turkish,
published only in September 1902, describing the ‘Conquest of Constantinople and
the establishment of the Turks in Europe.’ Its author is Achmed Muktar Pasha. It is
especially valuable as containing many quotations from Turkish authors who are
inaccessible to Europeans.
459 Barbaro, p. 56.
460 Crit. lvi.
461 Crit. lxiii.
462 The Horaia Gate occupied the site of the present Stamboul Custom House.
The Validé Mosque, at the end of the present outer bridge, is built on part of the
Jewish quarter. See the subject fully discussed by Professor van Millingen, p. 221
and elsewhere.
463 Leonard, 99; Phrantzes, 287.
464 Barbaro, pp. 55, 56.
465 The Moscovite, xxv. The whole chapter is full of improbable statements.
466 Ch. lxi.
467 Barbaro, p. 55.
468 Thyselii Expugnatio, ch. xxvi.
469 Phrantzes, p. 291.
470 P. 57.
471 The Capture of Constantinople, from the Taj-ut-Tavarikh by Khodja Sad-ud-
din. Translated by E. J. W. Gibb, p. 29.
472 Phrantzes, 287. Professor van Millingen (p. 189) believes that these towers
were a little to the south of the present Seraglio Lighthouse. One of them had an
interesting inscription, stating that it was built by the emperor Basil in 1024.
473 Another version of Tetaldi’s Informacion calls the galleys in question Venetian
(Dethier, p. 905).
474 Crit. ch. lxiii.
475 Barbaro, p. 57.

476 οὗ ἔσωθεν τῶν ἀδύτων καὶ ἄνωθεν τῶν θυσιαστηρίων καὶ τραπέζων ἤσθιον
καὶ ἔπινον καὶ τὰς ἀσελγεῖς γνώμας καὶ ὀρέξεις αὐτῶν μετὰ γυναικῶν καὶ παρθένων
καὶ παίδων ἐπάνωθεν ἐποίουν καὶ ἔπραττον. Phrantzes, p. 290.
477 Crit. xlii.
478 Ducas, xlii.: βιβλία ὑπὲρ ἀριθμόν.
479 P. 31. Khodja Sad-ud-din, translated by E. J. W. Gibb.
480 Report of Superior of Franciscans. He was present at the siege and arrived at
Bologna July 4, 1453.
481 Crit. lxvii. The Superior of the Franciscans reported that three thousand men
were killed on both sides on May 29. Probably we shall not be far wrong in saying
that between three and four thousand were killed on May 29 on the Christian side
and fifty thousand made prisoners.
482 Barbaro and Ducas.
483 Barbaro pretends, indeed, that they were the victims of a trick on the part of
the Genoese, who wished to secure their own safety by seizing their ships and
delivering them to Mahomet. His story, like everything else he says about the
Genoese, may well be doubted.
484 A portion of the chain which formed part of the boom is now in the narthex of
St. Irene. Its links average about eighteen inches long.
485 Tetaldi states that the Turks captured a Genoese ship and from thirteen to
sixteen others.
486 Ducas says five.
487 Crit. lxvii.
488 Ibid. lxiii.
489 About three fourths of the sea-walls were taken down. The remaining fourth
was spared, and a portion of them near Azap Capou still remains.
490 Angeli Johannis Zachariae Potestatis Perae Epistola. Leonard, p. 100. Ducas
says that Mahomet had an inventory made of the property of those who had fled,
and gave the owners three months within which to return, failing which, it would
be confiscated.
491 Zorzo Dolfin, p. 1040. See also Sauli’s Colonia dei Genovesi in Galata, vol. ii. p.
172, and Von Hammer, vol. ii., where the treaty is given in full in the appendix.
Usually Dolfin’s narrative is taken from Leonard, but the paragraphs relating to the
capitulations are an exception. Dolfin uses the word Privilegio. The capitulations
are called at different times by different names: grants, concessions, privileges,
capitulations, or treaties. I have already pointed out, in the Fall of Constantinople,

that the system of ex-territoriality, under which, in virtue of capitulations,
foreigners resident in Turkey are always under the protection of their own laws, is
the survival of the system once general in the Roman empire. Of course it is
ridiculous to speak of the capitulations as having been wrongfully wrung from the
Turks by Western nations, and equally absurd to claim that their grant shows the
far-reaching policy of the Turks in their desire to attract foreign trade. The Turks
found the system of ex-territoriality in full force and maintained it, being unwilling,
as they still are, to allow Christians, whether their own subjects or foreigners, to
rank on an equality with Moslems.
492 Ducas makes the entry to Hagia Sophia on the 30th. Phrantzes and
Chalcondylas, on the 29th.
493 Cantemir, vol. ii. p. 45 (ed. Paris, 1743). He gives the Persian text.
494 Report of podestà; Philip the Armenian, p. 680; also Leonard, 101.
495 Riccherio (p. 967), whose narrative is singularly clear and readable. See also
the report of the Superior of the Franciscans.
496 Phrantzes, 385.
497 Ibid. p. 383: ἐν ᾧ δὴ χρόνῳ καὶ μηνὶ ἀνεῖλεν αὐτοχειρίᾳ τὸν φίλτατόν μου
υἱὸν Ἰωάννην ὁ ἀσεβέστατος καὶ ἀπηνέστατος ἀμηρᾶς, ὃς δῆθεν ἐβούλετο τὴν
ἀθέμιτον σοδομίαν πρᾶξαι κατὰ τοῦ παιδός.
498 Crit. lxxiii.
499 Ibid.
500 Ducas, p. 137: ἐμφάνισας αὐτὰς τῷ αἱμοβόρῳ θηρίῳ.
501 Phrantzes, 291.
502 Pusculus also is violently hostile to Notaras, and probably for the same
reason: because he would not accept the Union.
503 Ducas, 137.
504 Crit. (lxiii.) gives a different version. He states that he tried to pass as a Turk,
in which his knowledge of the Turkish language aided him: but that he was
recognised and flung himself from the walls. His head was cut off and carried to
the sultan, who had offered a great reward for his capture dead or alive.
505 Crit. lxiii and lxvii.
506 Ibid. lxvii.
507 Report, p. 940. The houses were empty and bore the marks of the reckless
ravages of a savage horde.
508 Crit. lxix.

509 Ducas, 142.
510 Crit. bk. ii. ch. i.
511 Von Hammer states that the walls were completely repaired in 1477, but gives
no authority (Histoire de l’empire ottoman, iii. 209). A valuable hint is obtained
from Knolles, who, writing his history of the Turks in 1610, says that ‘the two utter
walls with the whole space between them are now but slenderly maintained by the
Turks, lying full of earth and other rubbish’ (Knolles’s History, p. 341, 3rd ed.
1621). The lowest of the three walls has almost entirely disappeared except as to
the lower portion, which forms one of the sides of the foss. In the Lycus valley,
and even throughout the whole length of the landward walls, I think it is manifest
to an observer that only the Inner Wall has been repaired.
512 Crit. lxxiii.
513 Ibid. lxxiv.
514 Crit. lxxv.
515 Phrantzes, 304.
516 Crit. bk. ii. ch. i.
517 Crit. bk. ii. ch. ii.
518 Ecclesiastical and Civil Affairs after the Conquest, by Athanasius Comnenos
Hypsilantes, pp. 1, 2. The version of Phrantzes agrees with that given above. He
gives a full account of the usual procedure on the appointment of a patriarch and
confirms the statement that the Church of the Apostles was assigned to Gennadius
as an official residence. Subsequently it was taken from the Greeks, was destroyed
and replaced by a mosque built in honour of the conqueror and known as the
Mahmoudieh. The former patriarch, says Phrantzes, was dead.
519 Crit. bk. iii. ch. v.
520 Commentari di Theo. Spandugino Cantacusino.
521 All these illustrations are from book ii. of Critobulus.
522 Fallmerayer’s Geschichte des Kaiserthums von Trapezunt. Not only is this work
the great authority for the history of Trebizond, but Fallmerayer himself brought to
light the most valuable materials for its history. He was the discoverer in Venice of
the chronicles of Panaretos in the library of Cardinal Bessarion. Since Fallmerayer
wrote, the MS. of Critobulus has been discovered. In book iv. a full account is
given of the capture of Trebizond and the treatment of its emperors. Finlay’s
History of Trebizond is very good, but he wrote without seeing the account of
Critobulus.
523 iii. 302.

524 Crit. bk. iii. ch. xxi. and xxii.
525 Von Hammer, iii. 282.
526 i. 32.
527 Voyage au Levant par ordre du roy, 1630.
528 Turcorum Origo, p. 22.
529 This was Gentile Bellini, who arrived in Constantinople in 1479 and left at the
end of 1480. He was sent, at the request of the sultan, by the Doge of Venice.
530 Crit. bk. iv. ch. ix.
531 Ibid. bk. v. ch. x.
532 Crit. bk. v. ch. xi. It is possible that as some of the Latin writers spoke of the
Turks as Teucri, in the belief that they were the descendants of the Trojans,
Mahomet may have been under the same illusion.
533 Les Sultans Ottomans, par Halil Ganem, p. 129 (Paris, 1901).
534 Chalcondylas.
535 These and many other fictions of the like kind come from Spandugino and
Sansovino.
536 Zorzo Dolfin (p. 985) says: ‘E homo non dedito a libidine, sobrio, in tempo del
ramadan non vol aldir sobrieta; a nulla volupta, a nulla piacea e dedito saluo a
gloria.’ This is in striking contradiction with Barbaro’s account, which in describing
Mahomet says, ‘Che a un momento importantissimo alla vigilia della gran bataglia
s’inebriò col capedan pascia secondo la sua usanza.’ Barbaro’s narrative is written
immediately after the capture of the city, and, as usual, he is careless of the
accusations which he brings against the Turks or Genoese.
537 Zorzo Dolfin, p. 936.
538 Les Sultans Ottomans, pp. 150 and 125.
539 The fascination of the old Greek stories still continues even among the
poorest Greeks, and it is astonishing how generally they are known. I have often
heard old Greek women, unable to read or write, tell children Greek paramythia
which have evidently been handed down by oral tradition. A few years ago, in
travelling among the mountains of Bithynia, I came on Easter Monday to a Greek
village, far remote from any other, and away from all lines of communication,
where they were performing a miracle-play. The villagers, dressed in their best,
were all present as actors or spectators. The play itself was a curious mixture of
incidents in the life of Christ and of others—and these formed the largest part—
from Greek mythology. No one knew anything of its origin, and all the information
obtainable was that the play had always been performed on Easter Monday.

540 See Aristarchi’s (the Grand Logothete) papers on Photius in the Transactions
of the Greek Syllogos of Constantinople, and two volumes edited by him of that
patriarch’s sermons and homilies, published 1901.
541 Heeren, in his Essai sur les Croisades, p. 413, quoted in Hallam’s Middle Ages,
ascribes the loss of all the authors missing from the library of Photius to the Latin
capture. Probably the statement is too sweeping.
542 Gibbon, vol. vii. 116.
543 See H. F. Tozer’s article on ‘The Greek-speaking Population of Southern Italy,’
in Journal of Hellenic Studies, x. p. 99.
544 ‘Nemo est qui Graecas literas novit.’ Quoted in Hodius, De Graecis illustribus,
p. 8.
545 Hodius, De Graecis illust.
546 Hodius, p. 28.
547 Philelphi Epis. in 1451.
548 Filelfo died in 1481. Dethier gives the letter which he wrote to Mahomet
praying for the release of his mother-in-law, a prayer which was granted.
549 Das Schriftwesen im Mittelalter (Leipzig, 1875), pp. 392 etc.
550 Burckhardt’s Renaissance in Italy, p. 192.
551 Gibbon selects some examples to show the anti-christian character of the
classical enthusiasm. (1) At the Council of Florence, Gemistos Pletho said in
familiar conversation to George of Trebizond that in a short time mankind would
unanimously renounce the Gospel and the Koran for a religion similar to that of
the Gentiles (Leo Allatius). (2) Paul II. accused the principal members of the
Roman Academy of heresy, impiety, and paganism (Tiraboschi). I suspect the first
charge of being grossly exaggerated or invented, but the fact that such a
statement could be credited shows to what extent the classical reaction had gone.
552 It is curious that the non-progressive party in Oxford, who violently opposed
the introduction of the new studies, called themselves Trojans. Roper’s Life of Sir
T. More (ed. Hearne), p. 75. The archbishops of Chios and Pusculus invariably
describe the Turks as Teucri.
553 Exuviae sacrae Constantinopolitanae.
554 Ducas, xliii.
555 αἱ πλείους δὲ αὐτῶν, οὐ πρὸς ἀπόδοσιν μᾶλλον ἢ ὕβριν &c. Crit. ch. lxii.
556 Hodius, De Graecis illustribus.

557 Aeneas Sylvius, in 1454, before the diet of Frankfort says: ‘Quid de libris
dicam, qui illic erant innumerabiles, nondum Latinis cogniti?... Nunc ergo et
Homero et Pindaro et omnibus illustrioribus poetis secunda mors erit.’
558 One such at least still remains at Zeirek Jami.
559 Probably more manuscripts existing as rolls (the original volumen) than in
book form have disappeared. The Turks, for example, when they occupied Mount
Athos during the Greek revolution, found the rolls very convenient for making
haversacks. The books have perished mostly from neglect. The discovery by the
present bishop of Ismidt of the Teaching of the Twelve Apostles (Διδαχὴ τῶν
δώδεκα ἀποστόλων) in 1883, in the library of a monastery on the Golden Horn
bound up with other manuscripts, the first of which only was indexed, gives hope
that others of value may yet be found. The same remark applies to the recovery,
about six years ago, of the Purple MS. of the Gospels, known technically as Codex
N, and now at St. Petersburg.
560 The influence of Byzantine art upon the West does not fall within the limits of
my task. But every one interested in the subject is aware that during some
centuries its influence was dominant. In the composition of pictures as well as in
their drawing and treatment Western artists for a long time copied those of
Constantinople. In painting, Byzantine influence prevailed throughout Italy from
Justinian to the middle of the fourteenth century. Giotto, who died in 1336, was,
says Kugler, the first to abandon the Byzantine style. In the intervening centuries
the monasteries of Constantinople, Salonica, and Mount Athos were the central
ateliers of painting, and furnished the models for artistic activity to all Europe. The
mosaics in the church of San Vitale at Ravenna are magnificent illustrations of
what Byzantine art was in the time of Justinian. Those in Hagia Sophia, as well as
its general plan of colour-ornamentation, are still unsurpassed. Those of the Kahrié
Mosque belonging to the fourteenth century are interesting and show a deep
feeling for colour-combination as well as accuracy of drawing. Byzantine
architecture in like manner greatly influenced the builders of churches in Western
lands. The front view of St. Mark’s in Venice in the thirteenth century placed side
by side with that of the Kahrié Mosque at the present day shows that the plan of
the earlier one was familiar to the architect of the other, and, as has been pointed
out by an architect who has made a careful study of the two buildings, when St.
Mark’s differs from the Kahrié, the difference may be found in details reproduced
from another church in Constantinople, that of the Pantocrator. The resemblance
between St. Mark’s and the Kahrié illustrates Mr. Fergusson’s observations on the
decoration of the exteriors of Byzantine churches. He points out that while the
interior of Hagia Sophia is ‘the most perfect and most beautiful church which has
yet been erected by any Christian people,’ the exterior was never finished
(Fergusson’s History of Architecture, ii. 321). The Kahrié of to-day resembles St.
Mark’s of the thirteenth century before the exterior casing was added to it.

The question of the influence of Byzantine art and architecture on the West has
often been dealt with. For a list of books on the subject see Karl Krumbacher’s
Geschichte der byzantinischen Litteratur, pp. 1124–27.
561 Hallam’s Middle Ages, ch. vi.
562 Angeli Johannis Epistola, p. 62.
563 See, for example, Cuspinianus, De Turcorum Origine; the author was in the
employ of the emperor Maximilian I. and insists again and again on the necessity
of resisting the Turk and the certainty of being able to do so with success. Almost
every European traveller in Turkey during two centuries, beginning with La
Brocquière and Tetaldi, made similar representations.
564 One of the best illustrations of the degraded position assigned to woman in
Mahometan countries is found in the fact that the popular belief is that she has no
soul. The influence of such a belief is of course fatal to the progress of the race. I
am well aware that Khaireddin Pasha and other progressive Mahometans have
maintained that this belief is contrary to the teaching of the Koran, and that Mr.
Hughes and other well-informed students of the sacred writings of Islam agree in
this opinion. Still, my statement as to the popular belief is not affected by these
researches into the original teaching. It is not alleged that the houris of Paradise
are the representatives of earthly women. The sensual rewards promised to
faithful men are clear and unmistakeable. The rewards to women in the Koran
have to be searched for and are the result of interpretation. As a confirmation of
the truth of my statement I may refer to the interesting interview given by Sir
Edward Malet in Shifting Scenes (1901), p. 67. He describes a meeting which he
had with Tewfik, the Khedive of Egypt, at a very critical moment, when indeed the
latter’s life was in hourly danger. He represents Tewfik as saying: ‘Death does not
signify to me personally. Our religion prevents us from having any fear of death;
but it is different with our women. To them, you know, life is everything: their
existence ends here; they cry and weep and implore me to save them.’
As to the custom of repudiating a wife, two learned Moslems, one Turkish and the
other Indian, and both enlightened men, assure me that repudiation, though a
general custom, is contrary to the teaching of Islam, which only recognises
divorce. Both, however, admit that the practice is general, though they consider it
irreligious or—what is the same thing in the Sacred Law of Islam—illegal.
565 I may add here that the great value of Christian missions from the West in the
Turkish Empire, those of the Latin Church and of the American Protestant
Churches alike, lies not only in their educational work but still more in their holding
up to the members of the Eastern Churches higher standards of truthfulness and
morality. Their influence has been already very useful. They have kindled a desire
for instruction, and have infused new life in many of the members of the ancient
Churches. While Greeks, Bulgarians, and Armenians look with intense distrust on

any attempts to proselytise, they have all been awakened by these missions to the
necessity for education. Considering the means at their disposal, I think it may be
fairly said that no other people during the last half-century has done so much for
education as the Greeks. The desire of every Greek who makes money seems to
be to found a school in his native place. In Constantinople several large and
excellent institutions, both for boys and girls, exist, all of course unaided by the
Government, and in other cities of the Turkish empire like efforts have been made
by patriotic Greeks. In Bulgaria one of the first acts of the newly enfranchised
state was to establish an efficient system of education. The Armenians are not
behind either, and their efforts, perhaps to a greater extent than those of the
other two peoples mentioned, are directed to bringing their priests into line with
those of the West. In 1896 the American missionaries in Turkey met in a ‘summer
school’ on the island of Proti, near Constantinople; the late Armenian patriarch
visited them, and, having spent a day in listening to their discussion on questions
of teaching and Biblical scholarship, declared that he would be ready to sacrifice
his life if his own priests could have the advantage of such gatherings.
566 ‘Pontes qui ad moenia ducunt dirumpunt.’ Pusculus iv. 137.
567 Professor van Millingen’s Byzantine Constantinople, p. 96.
568 Esquisse Topographique, p. 25.
569 Critobulus, Book II. ch. i.
570 Knolles, History of the Turks, p. 341 (written in 1610, edition of 1621).
571 P. 28.
572 1078, Dethier’s edition.
573 Byzantine Constantinople, p. 96. In the same manner Dethier, commenting on
Pusculus, iv. line 169, says: ‘Pseudoporta Charsaca vel Pempti omnium celeberrima
et in fortificatione calx Achilles erat. Hic enim ab utra parte, nempe a Porta
Polyandrii [Adrianople Gate] et a Porta Sancti Romani in vallem Lyci linea recta
murus descendit, idque contra omnem legem artis fortificationum.’
574 The Anonymous Chronicle, in verse, of the Latin Capture (edited by Joseph
Mueller and Dethier), line 390.
575 Threnos, 610–613.
576 Dethier and the elder Mordtmann considered (in error, as the learned son of
the latter and Professor van Millingen agree) that they had proved that the
Pempton was the Chariseus. See, in addition to the sentence just quoted from the
Threnos, the archaeological map of the Greek Syllogos and also Dethier’s note on
Pusculus, iv. line 172.
577 Ch. xxiii.: πρὸς ταῖς καλουμέναις πύλαις τοῦ Ῥωμανοῦ.

Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.
More than just a book-buying platform, we strive to be a bridge
connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.
Join us on a journey of knowledge exploration, passion nurturing, and
personal growth every day!
ebookbell.com

Intelligent Methods Systems And Applications In Computing Communications And Control Simona Dzitac

About This Presentation

Slide Content

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

Intelligent Methods Systems And Applications In Computing Communications And Control Simona Dzitac

About This Presentation

Slide Content

Slide 1

Slide 2

Slide 3

Slide 5

Slide 6

Slide 7

Slide 8

Slide 9

Slide 10

Slide 11

Slide 12

Slide 13

Slide 14

Slide 15

Slide 16

Slide 17

Slide 18

Slide 19

Slide 20

Slide 21

Slide 22

Slide 23

Slide 24

Slide 25

Slide 26

Slide 27

Slide 28

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

Slide 34

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Slide 40

Slide 41

Slide 42

Slide 43

Slide 44

Slide 45

Slide 46

Slide 47

Slide 48

Slide 49

Slide 50

Slide 51

Slide 52

Slide 53

Slide 54

Slide 55

Slide 56

Slide 57

Slide 58

Slide 59

Slide 60

Slide 61

Slide 62

Slide 63

Slide 64

Slide 65

Slide 66

Slide 67

Slide 68

Slide 69

Slide 70

Slide 71

Slide 72

Slide 73

Slide 74

Slide 75

Slide 76

Slide 77

Slide 78